Hybrid Arima - LSTM code - Google Trends¶
The hybrid ARIMA-LSTM model is open to a variety of experimentation. For ideal performance, a balance must be reached between the levels of volatility that work best for the ARIMA and LSTM models. Using shorter MA periods that result in a non-mesokurtic distribution may achieve a better volatility balance between models.
Import Libraries¶
In [1]:
import pandas as pd
pd.set_option('display.max_rows', 500)
import timeit
In [2]:
!pip install -q -U keras-tuner
|████████████████████████████████| 98 kB 3.9 MB/s
In [3]:
import keras_tuner as kt
In [4]:
!pip install pmdarima
Collecting pmdarima
Downloading pmdarima-1.8.4-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl (1.4 MB)
|████████████████████████████████| 1.4 MB 7.2 MB/s
Requirement already satisfied: numpy>=1.19.3 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.19.5)
Collecting statsmodels!=0.12.0,>=0.11
Downloading statsmodels-0.13.1-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (9.8 MB)
|████████████████████████████████| 9.8 MB 54.3 MB/s
Requirement already satisfied: pandas>=0.19 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.1.5)
Requirement already satisfied: setuptools!=50.0.0,>=38.6.0 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (57.4.0)
Requirement already satisfied: Cython!=0.29.18,>=0.29 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (0.29.24)
Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.1.0)
Requirement already satisfied: urllib3 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.24.3)
Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.4.1)
Requirement already satisfied: scikit-learn>=0.22 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.0.1)
Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.19->pmdarima) (2018.9)
Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.19->pmdarima) (2.8.2)
Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas>=0.19->pmdarima) (1.15.0)
Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.22->pmdarima) (3.0.0)
Requirement already satisfied: patsy>=0.5.2 in /usr/local/lib/python3.7/dist-packages (from statsmodels!=0.12.0,>=0.11->pmdarima) (0.5.2)
Installing collected packages: statsmodels, pmdarima
Attempting uninstall: statsmodels
Found existing installation: statsmodels 0.10.2
Uninstalling statsmodels-0.10.2:
Successfully uninstalled statsmodels-0.10.2
Successfully installed pmdarima-1.8.4 statsmodels-0.13.1
In [5]:
import pmdarima
In [6]:
url = 'https://launchpad.net/~mario-mariomedina/+archive/ubuntu/talib/+files'
!wget $url/libta-lib0_0.4.0-oneiric1_amd64.deb -qO libta.deb
!wget $url/ta-lib0-dev_0.4.0-oneiric1_amd64.deb -qO ta.deb
!dpkg -i libta.deb ta.deb
!pip install ta-lib
import talib
Selecting previously unselected package libta-lib0.
(Reading database ... 155222 files and directories currently installed.)
Preparing to unpack libta.deb ...
Unpacking libta-lib0 (0.4.0-oneiric1) ...
Selecting previously unselected package ta-lib0-dev.
Preparing to unpack ta.deb ...
Unpacking ta-lib0-dev (0.4.0-oneiric1) ...
Setting up libta-lib0 (0.4.0-oneiric1) ...
Setting up ta-lib0-dev (0.4.0-oneiric1) ...
Processing triggers for man-db (2.8.3-2ubuntu0.1) ...
Processing triggers for libc-bin (2.27-3ubuntu1.3) ...
/sbin/ldconfig.real: /usr/local/lib/python3.7/dist-packages/ideep4py/lib/libmkldnn.so.0 is not a symbolic link
Collecting ta-lib
Downloading TA-Lib-0.4.22.tar.gz (268 kB)
|████████████████████████████████| 268 kB 8.8 MB/s
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing wheel metadata ... done
Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from ta-lib) (1.19.5)
Building wheels for collected packages: ta-lib
Building wheel for ta-lib (PEP 517) ... done
Created wheel for ta-lib: filename=TA_Lib-0.4.22-cp37-cp37m-linux_x86_64.whl size=1465648 sha256=6864bc894fca760ea7a1568a10d29f53fda00cad2672992e2a9901b7a32c8f74
Stored in directory: /root/.cache/pip/wheels/7b/63/a9/144081748d9c4f0a09b4670c7b3c414bcb33ff97f0724c753a
Successfully built ta-lib
Installing collected packages: ta-lib
Successfully installed ta-lib-0.4.22
In [7]:
import tensorflow
import statsmodels.tsa.api
import keras
import sklearn
In [8]:
from tensorflow.keras.optimizers import Adam
from tensorflow.keras import Sequential
from tensorflow.keras.layers import Dense, LSTM, Dropout, Bidirectional,BatchNormalization, Embedding, TimeDistributed, LeakyReLU, GRU
from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau
In [9]:
from keras.models import Sequential, load_model
from keras.layers import Dense, LSTM, Activation, Dropout
from keras import backend as K
from keras.utils.generic_utils import get_custom_objects
from keras.callbacks import ModelCheckpoint,EarlyStopping
from keras.regularizers import l1_l2
In [10]:
import math
In [11]:
from statsmodels.tsa.api import VAR
from statsmodels.tsa.statespace.varmax import VARMAX,VARMAXResults
In [12]:
from sklearn.metrics import mean_squared_error, mean_absolute_percentage_error, mean_absolute_error
from sklearn.preprocessing import MinMaxScaler
from sklearn.model_selection import train_test_split
In [13]:
from matplotlib import pyplot
In [14]:
import json
import datetime
import pandas as pd
import numpy as np
import os
from scipy.stats import kurtosis
import pmdarima as pm
from pmdarima import auto_arima
from talib import abstract
import json
import matplotlib.pyplot as plt
# plt.rcParams.update({'font.size': 16})
from matplotlib.pyplot import figure
from numpy import array
from numpy import hstack
from keras.models import Sequential
from keras.layers import LSTM
from keras.layers import Dense
from keras.layers import RepeatVector
from keras.layers import TimeDistributed
In [15]:
from keras.utils.generic_utils import get_custom_objects
from tensorflow.keras.utils import plot_model
In [16]:
import warnings
from statsmodels.tools.sm_exceptions import ConvergenceWarning
warnings.simplefilter('ignore', ConvergenceWarning)
Load Data¶
In [2]:
from google.colab import drive
drive.mount('/content/drive')
Mounted at /content/drive
In [ ]:
cd drive/MyDrive/Stock price prediction/
In [18]:
cd drive/MyDrive/Stock price prediction/Generated datasets
/content/drive/.shortcut-targets-by-id/1IaGjVBlTspxI2CHSrxfYnaiYvsaG0pHs/Stock price prediction/Generated datasets
In [19]:
df = pd.read_csv("FULL_Data_google_COVID_bull_bear.csv",parse_dates=[0])
df.tail(10)
Out[19]:
| Unnamed: 0 | Unnamed: 0.1 | Unnamed: 0.1.1 | Unnamed: 0.1.1.1 | Open | High | Low | Close | Adj Close | Volume | MA7 | MA21 | MACD | 20SD | upper_band | lower_band | EMA | logmomentum | absolute of 3 comp | angle of 3 comp | absolute of 6 comp | angle of 6 comp | absolute of 9 comp | angle of 9 comp | Date | search | COVID positiveIncrease | COVID deathIncrease | bull score | bear score | fourier bull 10 | fourier bull 30 | fourier bear 10 | fourier bear 30 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1592 | 1592 | 1781 | 1781 | 1781 | 150.199997 | 151.429993 | 150.059998 | 150.809998 | 150.809998 | 56787900.0 | 150.565717 | 148.423811 | -1.137777 | 2.817933 | 154.059677 | 142.787944 | 150.767809 | 5.009368 | 93.428749 | -0.061228 | 100.779503 | -0.039111 | 103.599003 | -0.022436 | 2021-11-09 | 19 | 112313 | 1258 | 0.119141 | 0.111328 | NaN | NaN | NaN | NaN |
| 1593 | 1593 | 1782 | 1782 | 1782 | 150.020004 | 150.130005 | 147.850006 | 147.919998 | 147.919998 | 65187100.0 | 150.417145 | 148.729049 | -1.236913 | 2.144358 | 153.017766 | 144.440332 | 148.869268 | 4.989888 | 92.922909 | -0.061683 | 99.694365 | -0.039762 | 101.872301 | -0.022657 | 2021-11-10 | 19 | 80301 | 1470 | 0.154297 | 0.109375 | NaN | NaN | NaN | NaN |
| 1594 | 1594 | 1783 | 1783 | 1783 | 148.960007 | 149.429993 | 147.679993 | 147.869995 | 147.869995 | 41000000.0 | 150.110001 | 149.060477 | -1.165047 | 1.767475 | 152.595428 | 145.525526 | 148.203086 | 4.989548 | 92.416471 | -0.062129 | 98.604584 | -0.040391 | 100.137594 | -0.022839 | 2021-11-11 | 19 | 94975 | 1662 | 0.102845 | 0.126915 | NaN | NaN | NaN | NaN |
| 1595 | 1595 | 1784 | 1784 | 1784 | 148.429993 | 150.399994 | 147.479996 | 149.990005 | 149.990005 | 63632600.0 | 149.895715 | 149.357144 | -0.869308 | 1.420732 | 152.198608 | 146.515681 | 149.394365 | 5.003879 | 91.909483 | -0.062566 | 97.510555 | -0.040998 | 98.396260 | -0.022980 | 2021-11-12 | 19 | 55499 | 797 | 0.157277 | 0.080595 | NaN | NaN | NaN | NaN |
| 1596 | 1596 | 1785 | 1785 | 1785 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | 2021-11-13 | 19 | 146529 | 2505 | 0.139459 | 0.083243 | NaN | NaN | NaN | NaN |
| 1597 | 1597 | 1786 | 1786 | 1786 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | 2021-11-14 | 19 | 40964 | 479 | 0.151261 | 0.100840 | NaN | NaN | NaN | NaN |
| 1598 | 1598 | 1787 | 1787 | 1787 | 150.369995 | 151.880005 | 149.429993 | 150.000000 | 150.000000 | 59222800.0 | 149.758571 | 149.602859 | -0.907641 | 1.229694 | 152.062246 | 147.143471 | 149.798122 | 5.003946 | 91.401994 | -0.062993 | 96.412672 | -0.041581 | 96.649685 | -0.023077 | 2021-11-15 | 22 | 30290 | 148 | 0.136737 | 0.109389 | NaN | NaN | NaN | NaN |
| 1599 | 1599 | 1788 | 1788 | 1788 | 149.940002 | 151.490005 | 149.339996 | 151.000000 | 151.000000 | 59256200.0 | 149.718571 | 149.814763 | -0.791320 | 1.236243 | 152.287250 | 147.342277 | 150.599374 | 5.010635 | 90.894052 | -0.063410 | 95.311334 | -0.042140 | 94.899260 | -0.023130 | 2021-11-16 | 22 | 138962 | 1294 | 0.135531 | 0.115385 | NaN | NaN | NaN | NaN |
| 1600 | 1600 | 1789 | 1789 | 1789 | 151.000000 | 155.000000 | 150.990005 | 153.490005 | 153.490005 | 88807000.0 | 150.154286 | 150.040002 | -0.657719 | 1.467121 | 152.974245 | 147.105759 | 152.526461 | 5.027099 | 90.385704 | -0.063817 | 94.206941 | -0.042673 | 93.146378 | -0.023135 | 2021-11-17 | 22 | 87626 | 1290 | 0.100870 | 0.126957 | NaN | NaN | NaN | NaN |
| 1601 | 1601 | 1790 | 1790 | 1790 | 153.710007 | 158.669998 | 153.050003 | 157.869995 | 157.869995 | 137659100.0 | 151.162857 | 150.450002 | -0.609656 | 2.267825 | 154.985653 | 145.914351 | 156.088817 | 5.055417 | 89.877000 | -0.064214 | 93.099895 | -0.043179 | 91.392433 | -0.023090 | 2021-11-18 | 22 | 111404 | 1637 | 0.145098 | 0.121569 | NaN | NaN | NaN | NaN |
In [ ]:
cd ..
In [24]:
cd Archana - LSTM Hybrid/Outputs/Gtrends
/content/drive/.shortcut-targets-by-id/1IaGjVBlTspxI2CHSrxfYnaiYvsaG0pHs/Stock price prediction/Archana - LSTM Hybrid/Outputs/Gtrends
In [25]:
pd.to_datetime(df[np.isnan(df.Close)==True]['Date']).dt.day_name().head(5)
Out[25]:
0 Saturday 1 Sunday 3 Tuesday 7 Saturday 8 Sunday Name: Date, dtype: object
In [26]:
len(pd.to_datetime(df[np.isnan(df.Close)==True]['Date']).dt.day_name())
Out[26]:
497
In [27]:
len(df)
Out[27]:
1602
In [28]:
len(df) - len(pd.to_datetime(df[np.isnan(df.Close)==True]['Date']).dt.day_name())
Out[28]:
1105
In [29]:
df.dropna(inplace=True)
len(df)
Out[29]:
1080
In [30]:
pd.to_datetime(df[np.isnan(df.Close)==True]['Date']).dt.day_name()
Out[30]:
Series([], Name: Date, dtype: object)
In [31]:
df.head(5)
Out[31]:
| Unnamed: 0 | Unnamed: 0.1 | Unnamed: 0.1.1 | Unnamed: 0.1.1.1 | Open | High | Low | Close | Adj Close | Volume | MA7 | MA21 | MACD | 20SD | upper_band | lower_band | EMA | logmomentum | absolute of 3 comp | angle of 3 comp | absolute of 6 comp | angle of 6 comp | absolute of 9 comp | angle of 9 comp | Date | search | COVID positiveIncrease | COVID deathIncrease | bull score | bear score | fourier bull 10 | fourier bull 30 | fourier bear 10 | fourier bear 30 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | 2 | 191 | 191 | 191 | 36.220001 | 36.325001 | 35.775002 | 35.875000 | 34.054882 | 57111200.0 | 36.173571 | 36.751904 | 0.303356 | 0.960520 | 38.672945 | 34.830864 | 35.924548 | 3.551770 | 38.458011 | 0.046984 | 29.704545 | 0.102857 | 43.304973 | -0.053955 | 2017-07-03 | 15 | 0 | 0 | 0.666667 | 0.000000 | 0.142778 | 0.146810 | 0.100537 | 0.099251 |
| 4 | 4 | 193 | 193 | 193 | 35.922501 | 36.197498 | 35.680000 | 36.022499 | 34.194897 | 86278400.0 | 36.095357 | 36.634762 | 0.328795 | 0.852735 | 38.340231 | 34.929292 | 35.989849 | 3.555991 | 38.240991 | 0.049445 | 29.954520 | 0.099254 | 43.438321 | -0.053936 | 2017-07-05 | 15 | 0 | 0 | 0.400000 | 0.000000 | 0.144487 | 0.145833 | 0.100630 | 0.096361 |
| 5 | 5 | 194 | 194 | 194 | 35.755001 | 35.875000 | 35.602501 | 35.682499 | 33.872143 | 96515200.0 | 35.984999 | 36.495238 | 0.346702 | 0.677629 | 37.850495 | 35.139980 | 35.784949 | 3.546235 | 38.027974 | 0.051918 | 30.209839 | 0.095602 | 43.557403 | -0.053820 | 2017-07-06 | 15 | 0 | 0 | 0.142857 | 0.142857 | 0.145346 | 0.145164 | 0.100672 | 0.094761 |
| 6 | 6 | 195 | 195 | 195 | 35.724998 | 36.187500 | 35.724998 | 36.044998 | 34.216255 | 76806800.0 | 36.001071 | 36.362023 | 0.387422 | 0.387634 | 37.137291 | 35.586756 | 35.958315 | 3.556633 | 37.818962 | 0.054401 | 30.470232 | 0.091907 | 43.662260 | -0.053608 | 2017-07-07 | 15 | 0 | 0 | 0.333333 | 0.000000 | 0.146208 | 0.144377 | 0.100711 | 0.093072 |
| 9 | 9 | 198 | 198 | 198 | 36.027500 | 36.487499 | 35.842499 | 36.264999 | 34.425095 | 84362400.0 | 35.973571 | 36.243809 | 0.388315 | 0.308042 | 36.859893 | 35.627725 | 36.162771 | 3.562891 | 37.613953 | 0.056893 | 30.735430 | 0.088177 | 43.752965 | -0.053302 | 2017-07-10 | 14 | 0 | 0 | 0.000000 | 0.000000 | 0.148802 | 0.141354 | 0.100808 | 0.087587 |
In [32]:
stock_col= list(df.columns)
stock_col = stock_col[4:len(stock_col)]
In [33]:
dataset_final = df[stock_col]
In [34]:
dataset_final.head(5)
Out[34]:
| Open | High | Low | Close | Adj Close | Volume | MA7 | MA21 | MACD | 20SD | upper_band | lower_band | EMA | logmomentum | absolute of 3 comp | angle of 3 comp | absolute of 6 comp | angle of 6 comp | absolute of 9 comp | angle of 9 comp | Date | search | COVID positiveIncrease | COVID deathIncrease | bull score | bear score | fourier bull 10 | fourier bull 30 | fourier bear 10 | fourier bear 30 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | 36.220001 | 36.325001 | 35.775002 | 35.875000 | 34.054882 | 57111200.0 | 36.173571 | 36.751904 | 0.303356 | 0.960520 | 38.672945 | 34.830864 | 35.924548 | 3.551770 | 38.458011 | 0.046984 | 29.704545 | 0.102857 | 43.304973 | -0.053955 | 2017-07-03 | 15 | 0 | 0 | 0.666667 | 0.000000 | 0.142778 | 0.146810 | 0.100537 | 0.099251 |
| 4 | 35.922501 | 36.197498 | 35.680000 | 36.022499 | 34.194897 | 86278400.0 | 36.095357 | 36.634762 | 0.328795 | 0.852735 | 38.340231 | 34.929292 | 35.989849 | 3.555991 | 38.240991 | 0.049445 | 29.954520 | 0.099254 | 43.438321 | -0.053936 | 2017-07-05 | 15 | 0 | 0 | 0.400000 | 0.000000 | 0.144487 | 0.145833 | 0.100630 | 0.096361 |
| 5 | 35.755001 | 35.875000 | 35.602501 | 35.682499 | 33.872143 | 96515200.0 | 35.984999 | 36.495238 | 0.346702 | 0.677629 | 37.850495 | 35.139980 | 35.784949 | 3.546235 | 38.027974 | 0.051918 | 30.209839 | 0.095602 | 43.557403 | -0.053820 | 2017-07-06 | 15 | 0 | 0 | 0.142857 | 0.142857 | 0.145346 | 0.145164 | 0.100672 | 0.094761 |
| 6 | 35.724998 | 36.187500 | 35.724998 | 36.044998 | 34.216255 | 76806800.0 | 36.001071 | 36.362023 | 0.387422 | 0.387634 | 37.137291 | 35.586756 | 35.958315 | 3.556633 | 37.818962 | 0.054401 | 30.470232 | 0.091907 | 43.662260 | -0.053608 | 2017-07-07 | 15 | 0 | 0 | 0.333333 | 0.000000 | 0.146208 | 0.144377 | 0.100711 | 0.093072 |
| 9 | 36.027500 | 36.487499 | 35.842499 | 36.264999 | 34.425095 | 84362400.0 | 35.973571 | 36.243809 | 0.388315 | 0.308042 | 36.859893 | 35.627725 | 36.162771 | 3.562891 | 37.613953 | 0.056893 | 30.735430 | 0.088177 | 43.752965 | -0.053302 | 2017-07-10 | 14 | 0 | 0 | 0.000000 | 0.000000 | 0.148802 | 0.141354 | 0.100808 | 0.087587 |
Data Load for Experiment set 1 with Technical Indicators & google trends¶
In [35]:
stock_col= list(df.columns)
stock_col = stock_col[4:len(stock_col)-8]
dataset_final = df[stock_col]
dataset_final.head(5)
Out[35]:
| Open | High | Low | Close | Adj Close | Volume | MA7 | MA21 | MACD | 20SD | upper_band | lower_band | EMA | logmomentum | absolute of 3 comp | angle of 3 comp | absolute of 6 comp | angle of 6 comp | absolute of 9 comp | angle of 9 comp | Date | search | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | 36.220001 | 36.325001 | 35.775002 | 35.875000 | 34.054882 | 57111200.0 | 36.173571 | 36.751904 | 0.303356 | 0.960520 | 38.672945 | 34.830864 | 35.924548 | 3.551770 | 38.458011 | 0.046984 | 29.704545 | 0.102857 | 43.304973 | -0.053955 | 2017-07-03 | 15 |
| 4 | 35.922501 | 36.197498 | 35.680000 | 36.022499 | 34.194897 | 86278400.0 | 36.095357 | 36.634762 | 0.328795 | 0.852735 | 38.340231 | 34.929292 | 35.989849 | 3.555991 | 38.240991 | 0.049445 | 29.954520 | 0.099254 | 43.438321 | -0.053936 | 2017-07-05 | 15 |
| 5 | 35.755001 | 35.875000 | 35.602501 | 35.682499 | 33.872143 | 96515200.0 | 35.984999 | 36.495238 | 0.346702 | 0.677629 | 37.850495 | 35.139980 | 35.784949 | 3.546235 | 38.027974 | 0.051918 | 30.209839 | 0.095602 | 43.557403 | -0.053820 | 2017-07-06 | 15 |
| 6 | 35.724998 | 36.187500 | 35.724998 | 36.044998 | 34.216255 | 76806800.0 | 36.001071 | 36.362023 | 0.387422 | 0.387634 | 37.137291 | 35.586756 | 35.958315 | 3.556633 | 37.818962 | 0.054401 | 30.470232 | 0.091907 | 43.662260 | -0.053608 | 2017-07-07 | 15 |
| 9 | 36.027500 | 36.487499 | 35.842499 | 36.264999 | 34.425095 | 84362400.0 | 35.973571 | 36.243809 | 0.388315 | 0.308042 | 36.859893 | 35.627725 | 36.162771 | 3.562891 | 37.613953 | 0.056893 | 30.735430 | 0.088177 | 43.752965 | -0.053302 | 2017-07-10 | 14 |
In [36]:
# Set the date to datetime data
datetime_series = pd.to_datetime(dataset_final['Date'])
datetime_index = pd.DatetimeIndex(datetime_series.values)
dataset_final = dataset_final.set_index(datetime_index)
dataset_final = dataset_final.sort_values(by='Date')
dataset_final = dataset_final.drop(columns='Date')
dataset_final.head(5)
Out[36]:
| Open | High | Low | Close | Adj Close | Volume | MA7 | MA21 | MACD | 20SD | upper_band | lower_band | EMA | logmomentum | absolute of 3 comp | angle of 3 comp | absolute of 6 comp | angle of 6 comp | absolute of 9 comp | angle of 9 comp | search | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2017-07-03 | 36.220001 | 36.325001 | 35.775002 | 35.875000 | 34.054882 | 57111200.0 | 36.173571 | 36.751904 | 0.303356 | 0.960520 | 38.672945 | 34.830864 | 35.924548 | 3.551770 | 38.458011 | 0.046984 | 29.704545 | 0.102857 | 43.304973 | -0.053955 | 15 |
| 2017-07-05 | 35.922501 | 36.197498 | 35.680000 | 36.022499 | 34.194897 | 86278400.0 | 36.095357 | 36.634762 | 0.328795 | 0.852735 | 38.340231 | 34.929292 | 35.989849 | 3.555991 | 38.240991 | 0.049445 | 29.954520 | 0.099254 | 43.438321 | -0.053936 | 15 |
| 2017-07-06 | 35.755001 | 35.875000 | 35.602501 | 35.682499 | 33.872143 | 96515200.0 | 35.984999 | 36.495238 | 0.346702 | 0.677629 | 37.850495 | 35.139980 | 35.784949 | 3.546235 | 38.027974 | 0.051918 | 30.209839 | 0.095602 | 43.557403 | -0.053820 | 15 |
| 2017-07-07 | 35.724998 | 36.187500 | 35.724998 | 36.044998 | 34.216255 | 76806800.0 | 36.001071 | 36.362023 | 0.387422 | 0.387634 | 37.137291 | 35.586756 | 35.958315 | 3.556633 | 37.818962 | 0.054401 | 30.470232 | 0.091907 | 43.662260 | -0.053608 | 15 |
| 2017-07-10 | 36.027500 | 36.487499 | 35.842499 | 36.264999 | 34.425095 | 84362400.0 | 35.973571 | 36.243809 | 0.388315 | 0.308042 | 36.859893 | 35.627725 | 36.162771 | 3.562891 | 37.613953 | 0.056893 | 30.735430 | 0.088177 | 43.752965 | -0.053302 | 14 |
Train & test Dataset for Multistep Process¶
In [37]:
# Get features and target
X_value = pd.DataFrame(dataset_final.iloc[:, :])
y_value = pd.DataFrame(dataset_final.iloc[:, 3])
In [38]:
y_value.head(5)
Out[38]:
| Close | |
|---|---|
| 2017-07-03 | 35.875000 |
| 2017-07-05 | 36.022499 |
| 2017-07-06 | 35.682499 |
| 2017-07-07 | 36.044998 |
| 2017-07-10 | 36.264999 |
In [39]:
# Normalized the data
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
Out[39]:
MinMaxScaler(feature_range=(-1, 1))
In [40]:
X_scale_dataset = X_scaler.fit_transform(X_value)
y_scale_dataset = y_scaler.fit_transform(y_value)
In [41]:
X_scale_dataset.shape, y_scale_dataset.shape,
Out[41]:
((1080, 21), (1080, 1))
In [42]:
X_value.shape[1]
Out[42]:
21
N Steps Definition¶
In [43]:
n_steps_in = 3
n_features = X_value.shape[1] #19 features
n_steps_out = 1
In [44]:
# Reshape the data
'''Set the data input steps and output steps,
we use 30 days data to predict 1 day price here,
reshape it to (None, input_step, number of features) used for LSTM input'''
# Get X/y dataset
def get_X_y(X_data, y_data):
X = list()
y = list()
yc = list()
length = len(X_data)
for i in range(0, length, 1):
# pdb.set_trace()
X_value = X_data[i: i + n_steps_in][:, :]
# print('[',i,': ',i,' + ',n_steps_in,'][:, :]')
y_value = y_data[i + n_steps_in: i + (n_steps_in + n_steps_out)][:, 0]
# print('[',i,' + ',n_steps_in,': ',i,' + (',n_steps_in,' + ',n_steps_out,')][:, 0]')
yc_value = y_data[i: i + n_steps_in][:, :]
if len(X_value) == 3 and len(y_value) == 1:
X.append(X_value)
y.append(y_value)
yc.append(yc_value)
return np.array(X), np.array(y), np.array(yc)
In [45]:
# get the train test predict index
def predict_index(dataset, X_train, n_steps_in, n_steps_out):
# get the predict data (remove the in_steps days)
train_predict_index = dataset.iloc[n_steps_in : X_train.shape[0] + n_steps_in + n_steps_out - 1, :].index
test_predict_index = dataset.iloc[X_train.shape[0] + n_steps_in:, :].index
return train_predict_index, test_predict_index
In [46]:
def mean_absolute_percentage_error(actual, prediction):
actual = pd.Series(actual)
prediction = pd.Series(prediction)
return 100 * np.mean(np.abs((actual - prediction))/actual)
In [47]:
# Split train/test dataset
def split_train_test(data):
train_size = round(len(X) * 0.75)
data_train = data[0:train_size]
data_test = data[train_size:]
return data_train, data_test
In [48]:
# Get data and check shape
X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X)
y_train, y_test, = split_train_test(y)
yc_train, yc_test, = split_train_test(yc)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
In [49]:
# %% --------------------------------------- Save dataset -----------------------------------------------------------------
print('X shape: ', X.shape)
print('y shape: ', y.shape)
print('X_train shape: ', X_train.shape)
print('y_train shape: ', y_train.shape)
print('y_c_train shape: ', yc_train.shape)
print('X_test shape: ', X_test.shape)
print('y_test shape: ', y_test.shape)
print('y_c_test shape: ', yc_test.shape)
print('index_train shape:', index_train.shape)
print('index_test shape:', index_test.shape)
X shape: (1077, 3, 21) y shape: (1077, 1) X_train shape: (808, 3, 21) y_train shape: (808, 1) y_c_train shape: (808, 3, 1) X_test shape: (269, 3, 21) y_test shape: (269, 1) y_c_test shape: (269, 3, 1) index_train shape: (808,) index_test shape: (269,)
In [50]:
output_dim = y_train.shape[1]
output_dim
Out[50]:
1
In [51]:
df = dataset_final.copy()
In [52]:
df.rename(columns={'Date':'date','Open':'open','Low':'low','Close':'close','Volume':'volume','High':'high'}, inplace = True)
df.reset_index(drop=True,inplace=True)
In [53]:
df.head(1)
Out[53]:
| open | high | low | close | Adj Close | volume | MA7 | MA21 | MACD | 20SD | upper_band | lower_band | EMA | logmomentum | absolute of 3 comp | angle of 3 comp | absolute of 6 comp | angle of 6 comp | absolute of 9 comp | angle of 9 comp | search | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 36.220001 | 36.325001 | 35.775002 | 35.875 | 34.054882 | 57111200.0 | 36.173571 | 36.751904 | 0.303356 | 0.96052 | 38.672945 | 34.830864 | 35.924548 | 3.55177 | 38.458011 | 0.046984 | 29.704545 | 0.102857 | 43.304973 | -0.053955 | 15 |
In [54]:
# df.drop(['volume', 'MACD','20SD','logmomentum','absolute of 3 comp','angle of 3 comp','absolute of 6 comp','angle of 6 comp','absolute of 9 comp','angle of 9 comp'], axis='columns', inplace=True) # only keep columns that can help as residuals in Arima Hybrid
In [55]:
df.head(1)
Out[55]:
| open | high | low | close | Adj Close | volume | MA7 | MA21 | MACD | 20SD | upper_band | lower_band | EMA | logmomentum | absolute of 3 comp | angle of 3 comp | absolute of 6 comp | angle of 6 comp | absolute of 9 comp | angle of 9 comp | search | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 36.220001 | 36.325001 | 35.775002 | 35.875 | 34.054882 | 57111200.0 | 36.173571 | 36.751904 | 0.303356 | 0.96052 | 38.672945 | 34.830864 | 35.924548 | 3.55177 | 38.458011 | 0.046984 | 29.704545 | 0.102857 | 43.304973 | -0.053955 | 15 |
Train & Test Length¶
In [56]:
test_len = len(X_test)
In [57]:
train_len = len(X_train )
In [58]:
test_len, train_len
Out[58]:
(269, 808)
Kurtosis Review¶
In [59]:
# Initialize moving averages from Ta-Lib, store functions in dictionary
# talib_moving_averages = ['SMA', 'EMA', 'WMA', 'DEMA', 'KAMA', 'MIDPOINT', 'MIDPRICE', 'T3', 'TEMA', 'TRIMA'] remove midprice due to outputbeing univariate
talib_moving_averages = ['SMA', 'EMA', 'WMA', 'DEMA', 'KAMA', 'MIDPOINT', 'T3', 'TEMA', 'TRIMA']
functions = {}
for ma in talib_moving_averages:
functions[ma] = abstract.Function(ma)
# Determine kurtosis "K" values for MA period 4-30
kurtosis_results = {'period': []}
for i in range(4, 100): # 100
kurtosis_results['period'].append(i)
for ma in talib_moving_averages:
# Run moving average, remove last N days (used later for test data set), trim MA result to last 30 days
ma_output = functions[ma](df[:-test_len], i).tail(14)
# Determine kurtosis "K" value
k = kurtosis(ma_output, fisher=False)
# add to dictionary
if ma not in kurtosis_results.keys():
kurtosis_results[ma] = []
kurtosis_results[ma].append(k)
kurtosis_results = pd.DataFrame(kurtosis_results)
kurtosis_results.to_csv('kurtosis_results.csv')
In [60]:
kurtosis_results.head(5)
Out[60]:
| period | SMA | EMA | WMA | DEMA | KAMA | MIDPOINT | T3 | TEMA | TRIMA | |
|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 4 | 2.272452 | 2.652772 | 2.896972 | 3.800351 | 2.299585 | 2.171369 | 1.978458 | 4.609342 | 2.411225 |
| 1 | 5 | 1.839451 | 2.355815 | 2.481058 | 3.327525 | 1.841282 | 1.826597 | 1.640277 | 4.262302 | 1.994382 |
| 2 | 6 | 1.583886 | 2.159532 | 2.194320 | 2.945924 | 1.536136 | 1.605787 | 1.510972 | 3.878845 | 1.679710 |
| 3 | 7 | 1.461290 | 2.026758 | 1.990629 | 2.651927 | 1.506197 | 1.558096 | 1.514015 | 3.510432 | 1.486348 |
| 4 | 8 | 1.447516 | 1.935302 | 1.853935 | 2.429648 | 1.509566 | 1.621595 | 1.601580 | 3.184123 | 1.373337 |
Optimized periods¶
In [63]:
# Determine period with K closest to 3 +/-5%
optimized_period = {}
# https://pypi.org/project/TA-Lib/ determines the type of moving average to use
# https://pandas.pydata.org/pandas-docs/version/0.17.0/generated/pandas.DataFrame.at.html#pandas.DataFrame.at
for ma in talib_moving_averages:
difference = np.abs(kurtosis_results[ma] - 3)
df_arimahyb = pd.DataFrame({'difference': difference, 'period': kurtosis_results['period']})
df_arimahyb = df_arimahyb.sort_values(by=['difference'], ascending=True).reset_index(drop=True)
if df_arimahyb.at[0, 'difference'] < 3 * 0.05:
optimized_period[ma] = df_arimahyb.at[0, 'period']
else:
print(ma + ' is not viable, best K greater or less than 3 +/-5%')
print('\nOptimized periods:', optimized_period)
TRIMA is not viable, best K greater or less than 3 +/-5%
Optimized periods: {'SMA': 17, 'EMA': 51, 'WMA': 49, 'DEMA': 89, 'KAMA': 18, 'MIDPOINT': 14, 'T3': 19, 'TEMA': 9}
In [64]:
optimized_period
Out[64]:
{'DEMA': 89,
'EMA': 51,
'KAMA': 18,
'MIDPOINT': 14,
'SMA': 17,
'T3': 19,
'TEMA': 9,
'WMA': 49}
Simulation Keys¶
In [ ]:
simulation = {}
for ma in optimized_period:
print(ma)
print(functions[ma])
print ( int( optimized_period[ma]))
# if ma in ['EMA','WMA','DEMA','KAMA','MIDPOINT']:
# print(ma)
low_vol = df.apply(lambda c: functions[ma](c, timeperiod = int( optimized_period[ma])))
low_vol = low_vol.fillna(0)
high_vol = pd.DataFrame()
df2 = df.copy()
for i in df2.columns:
if i in low_vol.columns:
high_vol[i] = df2[i].subtract(low_vol[i], fill_value=0)
SMA
SMA([input_arrays], [timeperiod=30])
Simple Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
17
EMA
EMA([input_arrays], [timeperiod=30])
Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
51
WMA
WMA([input_arrays], [timeperiod=30])
Weighted Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
49
DEMA
DEMA([input_arrays], [timeperiod=30])
Double Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
89
KAMA
KAMA([input_arrays], [timeperiod=30])
Kaufman Adaptive Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
18
MIDPOINT
MIDPOINT([input_arrays], [timeperiod=14])
MidPoint over period (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 14
Outputs:
real
14
T3
T3([input_arrays], [timeperiod=5], [vfactor=0.7])
Triple Exponential Moving Average (T3) (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 5
vfactor: 0.7
Outputs:
real
19
TEMA
TEMA([input_arrays], [timeperiod=30])
Triple Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
9
In [ ]:
low_vol.tail(20)
Out[ ]:
| open | high | low | close | Adj Close | volume | MA7 | MA21 | MACD | 20SD | upper_band | lower_band | EMA | logmomentum | absolute of 3 comp | angle of 3 comp | absolute of 6 comp | angle of 6 comp | absolute of 9 comp | angle of 9 comp | search | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1060 | 140.200839 | 141.942909 | 138.524500 | 140.171495 | 139.966842 | 8.852448e+07 | 142.165478 | 146.699207 | 1.815578 | 4.572948 | 155.845103 | 137.553312 | 140.365562 | 4.935800 | 105.739092 | -0.047411 | 125.318767 | -0.018291 | 140.471430 | -0.008749 | 19.573385 |
| 1061 | 139.425914 | 141.705469 | 138.035200 | 140.698014 | 140.492650 | 8.620711e+07 | 141.528981 | 145.978836 | 2.115887 | 4.189393 | 154.357621 | 137.600050 | 140.587196 | 4.939545 | 105.263514 | -0.048037 | 124.464999 | -0.019222 | 139.335869 | -0.009472 | 19.632022 |
| 1062 | 140.773058 | 142.636405 | 139.932338 | 141.733666 | 141.526843 | 7.421445e+07 | 141.294887 | 145.298477 | 2.211018 | 3.647690 | 152.593858 | 138.003097 | 141.351509 | 4.946870 | 104.786174 | -0.048658 | 123.598217 | -0.020150 | 138.164839 | -0.010188 | 19.698090 |
| 1063 | 142.179695 | 143.266994 | 141.127848 | 142.249061 | 142.041527 | 6.519616e+07 | 141.224295 | 144.665584 | 2.093072 | 3.241276 | 151.148137 | 138.183031 | 141.949877 | 4.950518 | 104.307114 | -0.049275 | 122.718682 | -0.021074 | 136.959041 | -0.010898 | 19.763505 |
| 1064 | 142.253947 | 144.008334 | 141.546689 | 142.555532 | 142.347589 | 6.254214e+07 | 141.336839 | 144.184381 | 1.988881 | 2.884864 | 149.954110 | 138.414652 | 142.353647 | 4.952685 | 103.826381 | -0.049886 | 121.826667 | -0.021994 | 135.719217 | -0.011600 | 20.311676 |
| 1065 | 142.782738 | 143.732491 | 141.438660 | 142.125353 | 141.918068 | 6.542511e+07 | 141.385297 | 143.758659 | 1.774804 | 2.626682 | 149.012024 | 138.505294 | 142.201451 | 4.949632 | 103.344020 | -0.050491 | 120.922446 | -0.022909 | 134.446150 | -0.012293 | 20.671514 |
| 1066 | 142.153085 | 142.656915 | 140.466684 | 141.564232 | 141.357788 | 7.040262e+07 | 141.585336 | 143.387397 | 1.634667 | 2.376817 | 148.141030 | 138.633764 | 141.776638 | 4.945637 | 102.860075 | -0.051092 | 120.006305 | -0.023818 | 133.140665 | -0.012977 | 20.900131 |
| 1067 | 142.177201 | 143.194327 | 140.977156 | 142.610382 | 142.402435 | 6.948112e+07 | 141.933749 | 143.094536 | 1.573317 | 2.074153 | 147.242842 | 138.946230 | 142.332468 | 4.953023 | 102.374593 | -0.051687 | 119.078535 | -0.024722 | 131.803627 | -0.013650 | 21.038585 |
| 1068 | 143.009006 | 144.052615 | 142.286776 | 143.812497 | 143.602819 | 6.805244e+07 | 142.378675 | 142.879716 | 1.473333 | 1.874158 | 146.628032 | 139.131400 | 143.319154 | 4.961467 | 101.887619 | -0.052275 | 118.139433 | -0.025618 | 130.435938 | -0.014311 | 21.116168 |
| 1069 | 143.380322 | 145.547752 | 142.940349 | 145.397429 | 145.185452 | 7.592729e+07 | 142.902069 | 142.813890 | 1.447641 | 1.844159 | 146.502207 | 139.125573 | 144.704671 | 4.972505 | 101.399198 | -0.052858 | 117.189304 | -0.026508 | 129.038540 | -0.014959 | 21.153587 |
| 1070 | 145.337970 | 147.615882 | 144.980528 | 147.444584 | 147.229635 | 7.653090e+07 | 143.644287 | 142.961273 | 1.284466 | 2.010227 | 146.981728 | 138.940819 | 146.531280 | 4.986604 | 100.909377 | -0.053435 | 116.228458 | -0.027389 | 127.612408 | -0.015592 | 21.165321 |
| 1071 | 147.375283 | 149.163050 | 146.995423 | 148.921380 | 148.704294 | 6.811986e+07 | 144.553694 | 143.236380 | 0.961952 | 2.270386 | 147.777152 | 138.695607 | 148.124680 | 4.996737 | 100.418203 | -0.054006 | 115.257214 | -0.028261 | 126.158555 | -0.016211 | 21.161363 |
| 1072 | 148.656821 | 150.010875 | 148.071943 | 149.870634 | 149.652170 | 6.425222e+07 | 145.660163 | 143.530869 | 0.589081 | 2.556352 | 148.643574 | 138.418164 | 149.288649 | 5.003230 | 99.925720 | -0.054570 | 114.275894 | -0.029124 | 124.678027 | -0.016812 | 21.148490 |
| 1073 | 149.806550 | 150.715254 | 149.026204 | 149.977942 | 149.759331 | 6.069918e+07 | 146.862121 | 143.785380 | 0.135134 | 2.805932 | 149.397244 | 138.173516 | 149.748178 | 5.003989 | 99.431976 | -0.055128 | 113.284828 | -0.029977 | 123.171903 | -0.017396 | 21.131204 |
| 1074 | 149.937482 | 150.666013 | 149.022091 | 149.911667 | 149.693162 | 5.465321e+07 | 147.905162 | 144.001463 | -0.245163 | 3.045742 | 150.092948 | 137.909978 | 149.857170 | 5.003545 | 98.937018 | -0.055679 | 112.284350 | -0.030820 | 121.641290 | -0.017961 | 25.016406 |
| 1075 | 150.228161 | 151.254072 | 149.586503 | 150.104281 | 149.885502 | 5.602702e+07 | 148.803988 | 144.237215 | -0.571069 | 3.270011 | 150.777237 | 137.697192 | 150.021910 | 5.004835 | 98.440892 | -0.056223 | 111.274800 | -0.031650 | 120.087330 | -0.018506 | 27.455491 |
| 1076 | 150.328251 | 150.997797 | 149.591175 | 149.912656 | 149.694163 | 5.484778e+07 | 149.449021 | 144.548659 | -0.850904 | 3.458615 | 151.465890 | 137.631428 | 149.949074 | 5.003520 | 97.943645 | -0.056759 | 110.256524 | -0.032469 | 118.511190 | -0.019029 | 28.912854 |
| 1077 | 150.525566 | 152.430694 | 150.099878 | 151.531571 | 151.310718 | 7.580033e+07 | 150.032876 | 144.967153 | -0.975625 | 3.719924 | 152.407001 | 137.527305 | 151.004072 | 5.014296 | 97.445324 | -0.057289 | 109.229873 | -0.033274 | 116.914063 | -0.019528 | 29.716707 |
| 1078 | 149.301052 | 151.688142 | 148.723104 | 151.137179 | 150.916905 | 1.012990e+08 | 150.349418 | 145.413317 | -0.891585 | 3.905336 | 153.223988 | 137.602646 | 151.092810 | 5.011652 | 96.945977 | -0.057811 | 108.195203 | -0.034066 | 115.297171 | -0.020004 | 30.096629 |
| 1079 | 149.321425 | 151.018197 | 148.455004 | 150.396057 | 150.176865 | 9.262134e+07 | 150.424479 | 145.823313 | -0.852689 | 3.878291 | 153.579894 | 138.066731 | 150.628308 | 5.006660 | 96.445650 | -0.058325 | 107.152874 | -0.034844 | 113.661756 | -0.020453 | 27.283213 |
In [ ]:
high_vol.head(10)
Out[ ]:
| open | high | low | close | Adj Close | volume | MA7 | MA21 | MACD | 20SD | upper_band | lower_band | EMA | logmomentum | absolute of 3 comp | angle of 3 comp | absolute of 6 comp | angle of 6 comp | absolute of 9 comp | angle of 9 comp | search | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 36.220001 | 36.325001 | 35.775002 | 35.875000 | 34.054882 | 57111200.0 | 36.173571 | 36.751904 | 0.303356 | 0.960520 | 38.672945 | 34.830864 | 35.924548 | 3.551770 | 38.458011 | 0.046984 | 29.704545 | 0.102857 | 43.304973 | -0.053955 | 15.0 |
| 1 | 35.922501 | 36.197498 | 35.680000 | 36.022499 | 34.194897 | 86278400.0 | 36.095357 | 36.634762 | 0.328795 | 0.852735 | 38.340231 | 34.929292 | 35.989849 | 3.555991 | 38.240991 | 0.049445 | 29.954520 | 0.099254 | 43.438321 | -0.053936 | 15.0 |
| 2 | 35.755001 | 35.875000 | 35.602501 | 35.682499 | 33.872143 | 96515200.0 | 35.984999 | 36.495238 | 0.346702 | 0.677629 | 37.850495 | 35.139980 | 35.784949 | 3.546235 | 38.027974 | 0.051918 | 30.209839 | 0.095602 | 43.557403 | -0.053820 | 15.0 |
| 3 | 35.724998 | 36.187500 | 35.724998 | 36.044998 | 34.216255 | 76806800.0 | 36.001071 | 36.362023 | 0.387422 | 0.387634 | 37.137291 | 35.586756 | 35.958315 | 3.556633 | 37.818962 | 0.054401 | 30.470232 | 0.091907 | 43.662260 | -0.053608 | 15.0 |
| 4 | 36.027500 | 36.487499 | 35.842499 | 36.264999 | 34.425095 | 84362400.0 | 35.973571 | 36.243809 | 0.388315 | 0.308042 | 36.859893 | 35.627725 | 36.162771 | 3.562891 | 37.613953 | 0.056893 | 30.735430 | 0.088177 | 43.752965 | -0.053302 | 14.0 |
| 5 | 36.182499 | 36.462502 | 36.095001 | 36.382500 | 34.536625 | 79127200.0 | 36.039642 | 36.202738 | 0.372153 | 0.308860 | 36.820458 | 35.585018 | 36.309257 | 3.566217 | 37.412947 | 0.059392 | 31.005161 | 0.084416 | 43.829622 | -0.052901 | 14.0 |
| 6 | 36.467499 | 36.544998 | 36.205002 | 36.435001 | 34.586472 | 99538000.0 | 36.101071 | 36.206547 | 0.317572 | 0.295861 | 36.798268 | 35.614826 | 36.393086 | 3.567700 | 37.215939 | 0.061899 | 31.279154 | 0.080632 | 43.892360 | -0.052406 | 14.0 |
| 7 | 36.375000 | 37.122501 | 36.360001 | 36.942501 | 35.068211 | 100797600.0 | 36.253571 | 36.220595 | 0.322643 | 0.340687 | 36.901969 | 35.539221 | 36.759363 | 3.581920 | 37.022928 | 0.064410 | 31.557136 | 0.076830 | 43.941338 | -0.051818 | 14.0 |
| 8 | 36.992500 | 37.332500 | 36.832500 | 37.259998 | 35.369610 | 80528400.0 | 36.430357 | 36.266785 | 0.257925 | 0.410484 | 37.087753 | 35.445818 | 37.093120 | 3.590715 | 36.833908 | 0.066926 | 31.838833 | 0.073014 | 43.976744 | -0.051137 | 14.0 |
| 9 | 37.205002 | 37.724998 | 37.142502 | 37.389999 | 35.493000 | 95174000.0 | 36.674285 | 36.329523 | 0.184267 | 0.445597 | 37.220717 | 35.438330 | 37.291039 | 3.594294 | 36.648875 | 0.069445 | 32.123972 | 0.069192 | 43.998789 | -0.050365 | 16.0 |
Common Functions¶
In [65]:
def get_arima(dataframe,original_data, train_len, test_len):
# prepare train and test data
X_value = pd.DataFrame(dataframe.iloc[:, :])
y_value = pd.DataFrame(dataframe.iloc[:, 3])
X_train, X_test = split_train_test(X_value)
y_train, y_test = split_train_test(y_value)
yc_train,yc_test = split_train_test(original_data)
# y_train_ = y_train['close'].to_list()
# y_test_ = y_test['close'].to_list()
yc = yc_test.values.tolist()
y_train_list = y_train['close'].values.tolist()
y_test_list = y_test['close'].values.tolist()
# Initialize model
model = auto_arima(y_train_list,trace=True, error_action='ignore', start_p=1,start_q=1,max_p=3,max_q=3,d=3,
suppress_warnings=True,stepwise=True,seasonal=True)
print(model.summary())
# Determine model parameters
model.fit(y_train_list,disp= 0)
order = model.get_params()['order']
print('ARIMA order:', order, '\n')
# Genereate predictions
prediction = []
for i in range(len(y_test_list)):
model = pmdarima.ARIMA(order=order)
model.fit(y_train_list,disp= 0)
# print('working on', i+1, 'of', len(y_test), '-- ' + str(int(100 * (i + 1) / len(y_test))) + '% complete')
prediction.append(model.predict()[0])
y_train_list.append(y_test_list[i])
# Generate error data
mse = mean_squared_error(yc_test, prediction)
rmse = mse ** 0.5
# mape = mean_absolute_percentage_error(pd.Series(yc_test).values.tolist(), pd.Series(predictionte).values.tolist() )
mae = mean_absolute_error(pd.Series(yc_test).values.tolist() , pd.Series(prediction).values.tolist() )
return yc, prediction, mse, rmse, mae
In [66]:
def plot_train(simulation,SIM):
train_predict_index = np.load("index_train_appl.npy", allow_pickle=True)#Dates for train data
predict_result = pd.DataFrame()
for i in range(len(simulation[SIM]['final_tr']['prediction'])):
y_predict = pd.DataFrame(simulation[SIM]['final_tr']['prediction'][i], columns=["predicted_price"],
index=train_predict_index[i:i + output_dim])
predict_result = pd.concat([predict_result, y_predict], axis=1, sort=False)
#This is a dataframe with each column containing the predicted daily closing price
real_price = pd.DataFrame()
for i in range(len(simulation[SIM]['final_tr']['original'])):
y_train = pd.DataFrame(simulation[SIM]['final_tr']['original'][i], columns=["real_price"],
index=train_predict_index[i:i + output_dim])
real_price = pd.concat([real_price, y_train], axis=1, sort=False) #This is a dataframe with each column containing the real daily closing price
predict_result['predicted_mean'] = predict_result.mean(axis=1)#Adding a column with the daily predicted closing price value
real_price['real_mean'] = real_price.mean(axis=1)#Adding a column with the daily real closing price value
#
# Plot the predicted result
plt.figure(figsize=(16, 8))
plt.plot(real_price["real_mean"])
plt.plot(predict_result["predicted_mean"], color='r')
plt.xlabel("Date")
plt.ylabel("Stock price")
plt.legend(("Real price", "Predicted price"), loc="upper left", fontsize=16)
plt.title(f"The result of Training for Hybrid Arima LSTM with MA - {SIM} : {fileimg}",fontsize=20)
sf = fileimg+'_'+SIM+'Train Hybrid Arima LSTM Pred Out.png'
plt.savefig(sf,dpi='figure')
plt.show()
# Calculate RMSE
predicted = predict_result["predicted_mean"]
real = real_price["real_mean"]
RMSE = np.sqrt(mean_squared_error(predicted, real))
MSE = mean_squared_error(predicted, real)
MAE = mean_absolute_error(predicted, real)
print(f"----- Train RMSE for {SIM} -----", RMSE)
print(f"----- Train_MSE_LSTM for {SIM} -----", MSE)
print(f"----- Train MAE LSTM for {SIM} -----", MAE)
In [67]:
def plot_test(simulation, SIM):
test_predict_index = np.load("index_test_appl.npy", allow_pickle=True)#Dates for train data
# rescaled_real_y = y_scaler.inverse_transform(y_train)#Real closing price data
# rescaled_predicted_y = y_scaler.inverse_transform(train_yhat)#Predicted closing price data
predict_result = pd.DataFrame()
for i in range(len(simulation[SIM]['final']['prediction'])):
y_predict = pd.DataFrame(simulation[SIM]['final']['prediction'][i], columns=["predicted_price"],
index=test_predict_index[i:i + output_dim])
predict_result = pd.concat([predict_result, y_predict], axis=1, sort=False)#This is a dataframe with each column containing the predicted daily closing price
#
real_price = pd.DataFrame()
for i in range(len(simulation[SIM]['final']['original'])):
y_train = pd.DataFrame(simulation[SIM]['final']['original'][i], columns=["real_price"],
index=test_predict_index[i:i + output_dim])
real_price = pd.concat([real_price, y_train], axis=1, sort=False)#This is a dataframe with each column containing the real daily closing price
predict_result['predicted_mean'] = predict_result.mean(axis=1)#Adding a column with the daily predicted closing price value
real_price['real_mean'] = real_price.mean(axis=1)#Adding a column with the daily real closing price value
#
# Plot the predicted result
plt.figure(figsize=(16, 8))
plt.plot(real_price["real_mean"])
plt.plot(predict_result["predicted_mean"], color='r')
plt.xlabel("Date")
plt.ylabel("Stock price")
plt.legend(("Real price", "Predicted price"), loc="upper left", fontsize=16)
plt.title(f"The result of Testing for Hybrid Arima LSTM with MA - {SIM} : {fileimg}",fontsize=20)
sf = fileimg+'_'+SIM+'Test Hybrid Arima LSTM Pred Out.png'
plt.savefig(sf,dpi='figure')
plt.show()
# Calculate RMSE
predicted = predict_result["predicted_mean"]
real = real_price["real_mean"]
RMSE = np.sqrt(mean_squared_error(predicted, real))
MSE = mean_squared_error(predicted, real)
MAE = mean_absolute_error(predicted, real)
print(f"----- Test RMSE for {SIM}-----", RMSE)
print(f"----- Test_MSE_LSTM for {SIM}-----", MSE)
print(f"----- Test_MAE_LSTM for {SIM}-----", MAE)
In [68]:
def plot_train_high(simulation, SIM):
train_predict_index = np.load("index_test_appl.npy", allow_pickle=True)#Dates for train data
predict_result = pd.DataFrame()
for i in range(len(simulation[SIM]['high_vol']['prediction'])):
y_predict = pd.DataFrame(simulation[SIM]['high_vol']['prediction'][i], columns=["predicted_price"],
index=train_predict_index[i:i + output_dim])
predict_result = pd.concat([predict_result, y_predict], axis=1, sort=False)
#This is a dataframe with each column containing the predicted daily closing price
real_price = pd.DataFrame()
for i in range(len(simulation[SIM]['high_vol']['original'])):
y_train = pd.DataFrame(simulation[SIM]['high_vol']['original'][i], columns=["real_price"],
index=train_predict_index[i:i + output_dim])
real_price = pd.concat([real_price, y_train], axis=1, sort=False) #This is a dataframe with each column containing the real daily closing price
predict_result['predicted_mean'] = predict_result.mean(axis=1)#Adding a column with the daily predicted closing price value
real_price['real_mean'] = real_price.mean(axis=1)#Adding a column with the daily real closing price value
#
# Plot the predicted result
plt.figure(figsize=(16, 8))
plt.plot(real_price["real_mean"])
plt.plot(predict_result["predicted_mean"], color='r')
plt.xlabel("Date")
plt.ylabel("Stock price")
plt.legend(("Real price", "Predicted price"), loc="upper left", fontsize=16)
plt.title(f"The result of Training for {SIM}", fontsize=20)
plt.show()
# Calculate RMSE
predicted = predict_result["predicted_mean"]
real = real_price["real_mean"]
RMSE = np.sqrt(mean_squared_error(predicted, real))
MSE = mean_squared_error(predicted, real)
MAE = mean_absolute_error(predicted, real)
print(f"----- Individual LSTM RMSE for {SIM} -----", RMSE)
print(f"----- Individual LSTM_MSE_LSTM for {SIM} -----", MSE)
print(f"----- Individual LSTM MAE LSTM for {SIM} -----", MAE)
In [69]:
def plot_train_low(simulation , SIM):
train_predict_index = np.load("index_test_appl.npy", allow_pickle=True)#Dates for train data
predict_result = pd.DataFrame()
for i in range(len(simulation[SIM]['low_vol']['prediction'])):
y_predict = pd.DataFrame(simulation[SIM]['low_vol']['prediction'][i], columns=["predicted_price"],
index=train_predict_index[i:i + output_dim])
predict_result = pd.concat([predict_result, y_predict], axis=1, sort=False)
#This is a dataframe with each column containing the predicted daily closing price
real_price = pd.DataFrame()
for i in range(len(simulation[SIM]['low_vol']['original'])):
y_train = pd.DataFrame(simulation[SIM]['low_vol']['original'][i], columns=["real_price"],
index=train_predict_index[i:i + output_dim])
real_price = pd.concat([real_price, y_train], axis=1, sort=False) #This is a dataframe with each column containing the real daily closing price
predict_result['predicted_mean'] = predict_result.mean(axis=1)#Adding a column with the daily predicted closing price value
real_price['real_mean'] = real_price.mean(axis=1)#Adding a column with the daily real closing price value
#
# Plot the predicted result
plt.figure(figsize=(16, 8))
plt.plot(real_price["real_mean"])
plt.plot(predict_result["predicted_mean"], color='r')
plt.xlabel("Date")
plt.ylabel("Stock price")
plt.legend(("Real price", "Predicted price"), loc="upper left", fontsize=16)
plt.title(f"The result of Training for {SIM}", fontsize=20)
plt.show()
# Calculate RMSE
predicted = predict_result["predicted_mean"]
real = real_price["real_mean"]
RMSE = np.sqrt(mean_squared_error(predicted, real))
MSE = mean_squared_error(predicted, real)
MAE = mean_absolute_error(predicted, real)
print(f"-----Arima RMSE for {SIM} -----", RMSE)
print(f"----- Arima MSE for {SIM} -----", MSE)
print(f"----- Arima MAE for {SIM} -----", MAE)
In [70]:
import os
os.getcwd()
Out[70]:
'/content/drive/.shortcut-targets-by-id/1IaGjVBlTspxI2CHSrxfYnaiYvsaG0pHs/Stock price prediction/Archana - LSTM Hybrid/Outputs/Gtrends'
Univariate Arima Multistep MutiVariate LSTM Hybrid Model Experiment 1¶
In [ ]:
def get_lstm(data,original_data, train_len, test_len,img_file,ma ,lstm_len=3):
# prepare train and test data
X_value = pd.DataFrame(data.iloc[:, :])
y_value = pd.DataFrame(data.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
# Get data and check shape
X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X)
y_train, y_test, = split_train_test(y)
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
det = 20
input_dim = X_train.shape[1]#3
feature_size = X_train.shape[2]#24
output_dim = y_train.shape[1]#1
# Option 1
# Set up & fit LSTM RNN
model = Sequential()
model.add(LSTM(256, activation='relu', kernel_initializer='he_normal', input_shape=(input_dim, feature_size)))
model.add(Dense(units=64,activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(units=output_dim))
model.compile(optimizer=Adam(learning_rate = 0.001), loss='mse')
## Common code
callbacks = [
EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
fname1 = img_file+'.png'
tensorflow.keras.utils.plot_model(
model, to_file=fname1, show_shapes=True, show_dtype=False,
show_layer_names=True, expand_nested=False, dpi=96,
layer_range=None, show_layer_activations=False
)
history = model.fit(X_train, y_train, epochs=500, batch_size=int( optimized_period[ma]), verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# plot loss
fname2 = img_file+'-'+ma
plt.title(img_file+'-'+ma+' Loss')
plt.xlabel("Epochs")
plt.ylabel("Loss")
pyplot.plot(history.history['loss'], label='train')
pyplot.plot(history.history['val_loss'], label='validation')
pyplot.legend()
pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# # option 2
# model = Sequential()
# model.add(Bidirectional(LSTM(units= 128), input_shape=(input_dim, feature_size)))
# model.add(Dense(64))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(lr = 0.001), loss='mean_squared_error', metrics=['accuracy'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Option 3
# define custom activation
# cts().update({'double_tanh':Double_Tanh(double_tanh)})
# # Model Generation
# model = Sequential()
# #check https://machinelearningmastery.com/use-weight-regularization-lstm-networks-time-series-forecasting/
# model.add(LSTM(25, input_shape=(input_dim, feature_size), dropout=0.2, kernel_regularizer=l1_l2(0.00,0.00), bias_regularizer=l1_l2(0.00,0.00)))
# model.add(Dense(1))
# model.add(Activation(double_tanh))
# model.compile(loss='mean_squared_error', optimizer='adam', metrics=['mse', 'mae'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Option 4
# Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(units=lstm_len, return_sequences=True, input_shape=(x_train.shape[1], 1)))
# model.add(LSTM(units=int(lstm_len/2)))
# model.add(Dense(1, activation='sigmoid'))
# model.compile(loss='mean_squared_error', optimizer='adam')
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Generate predictions
predictiontr = model.predict(X_train, verbose=0)
predictiontr = y_scaler.inverse_transform(predictiontr).tolist()
outputtr = []
for i in range(len(predictiontr)):
outputtr.extend(predictiontr[i])
predictiontr = outputtr
# Generate error data
## replace with yc , xtest generated by new multistep method
mse_tr = mean_squared_error(y_train, predictiontr)
rmse_tr = mse_tr ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictiontr))
mae_tr = mean_absolute_error(y_train, pd.Series(predictiontr))
# Original_tr = pd.Series(yc_train)
Original_tr = y_scaler.inverse_transform(y_train).flatten().tolist()
predictionte = model.predict(X_test, verbose=0)
predictionte = (y_scaler.inverse_transform(predictionte)-det).tolist()
outputte = []
for i in range(len(predictionte)):
outputte.extend(predictionte[i])
predictionte = outputte
# Generate error data
mse_te = mean_squared_error(y_test, predictionte)
rmse_te = mse_te ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictionte))
mae_te = mean_absolute_error(y_test, pd.Series(predictionte))
# Original_te = pd.Series(yc_test)
Original_te = y_scaler.inverse_transform(y_test).flatten().tolist()
return Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,Original_te,predictionte, mse_te,rmse_te,mae_te
In [ ]:
if __name__ == '__main__':
start_time = timeit.default_timer()
simulation1 = {}
imgfile = 'Experiment1'
for ma in optimized_period:
print(ma)
print(functions[ma])
print ( int( optimized_period[ma]))
# if ma == 'SMA':
low_vol = df.apply(lambda c: functions[ma](c, timeperiod = int( optimized_period[ma])))
low_vol = low_vol.fillna(0)
low_vol_data = df['close']
high_vol = pd.DataFrame()
df2 = df.copy()
for i in df2.columns:
if i in low_vol.columns:
high_vol[i] = df2[i].subtract(low_vol[i], fill_value=0)
high_vol_data = df['close']
## *****************************************************
# Generate ARIMA and LSTM predictions
print('\nWorking on ' + ma + ' predictions')
try:
print('parameters used : ', train_len, test_len)
low_vol_Original, low_vol_prediction, low_vol_mse, low_vol_rmse,low_vol_mae = get_arima(low_vol,low_vol_data, train_len, test_len)
except:
print('ARIMA error, skipping to next MA type')
continue
Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,high_vol_Original, high_vol_prediction, high_vol_mse, high_vol_rmse,high_vol_mae, = get_lstm(high_vol,high_vol_data, train_len, test_len,imgfile,ma)
final_prediction_tr = df['close'].head(train_len).values + pd.Series(predictiontr) # ignoring first 3 steps
mse_ftr = mean_squared_error(df['close'].head(train_len).values,final_prediction_tr.values)
rmse_ftr = mse_ftr ** 0.5
mape_ftr = mean_absolute_percentage_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
mae_ftr = mean_absolute_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
final_prediction = pd.Series(low_vol_prediction[3:]) + pd.Series(high_vol_prediction)
mse = mean_squared_error(df['close'].tail(test_len).values,final_prediction.values)
rmse = mse ** 0.5
mape = mean_absolute_percentage_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
mae = mean_absolute_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
# Generate prediction accuracy
actual = df['close'].tail(test_len).values
result_1 = []
result_2 = []
for i in range(1, len(final_prediction)):
# Compare prediction to previous close price
if final_prediction[i] > actual[i-1] and actual[i] > actual[i-1]:
result_1.append(1)
elif final_prediction[i] < actual[i-1] and actual[i] < actual[i-1]:
result_1.append(1)
else:
result_1.append(0)
# Compare prediction to previous prediction
if final_prediction[i] > final_prediction[i-1] and actual[i] > actual[i-1]:
result_2.append(1)
elif final_prediction[i] < final_prediction[i-1] and actual[i] < actual[i-1]:
result_2.append(1)
else:
result_2.append(0)
accuracy_1 = np.mean(result_1)
accuracy_2 = np.mean(result_2)
simulation1[ma] = {'low_vol': {'original':list(low_vol_Original), 'prediction': list(low_vol_prediction) , 'mse': low_vol_mse,
'rmse': low_vol_rmse, 'mae' : low_vol_mae},
'high_vol': {'original':list(high_vol_Original),'prediction': list(high_vol_prediction), 'mse': high_vol_mse,
'rmse': high_vol_rmse, 'mae' : high_vol_mae},
'final_tr': {'original':df['close'].head(train_len).tolist(),'prediction': final_prediction_tr.values.tolist(), 'mse': mse_ftr,
'rmse': rmse_ftr, 'mae' : mae_ftr},
'final': {'original': df['close'].tail(test_len).tolist(), 'prediction': final_prediction.values.tolist(), 'mse': mse,
'rmse': rmse, 'mae': mae },
'accuracy': {'prediction vs close': accuracy_1, 'prediction vs prediction': accuracy_2}}
# save simulation data here as checkpoint
with open('simulation1_data.json', 'w') as fp:
json.dump(simulation1, fp)
for ma in simulation1.keys():
print('\n' + ma)
print('Prediction vs Close:\t\t' + str(round(100*simulation1[ma]['accuracy']['prediction vs close'], 2))
+ '% Accuracy')
print('Prediction vs Prediction:\t' + str(round(100*simulation1[ma]['accuracy']['prediction vs prediction'], 2))
+ '% Accuracy')
print('MSE:\t', simulation1[ma]['final']['mse'],
'\nRMSE:\t', simulation1[ma]['final']['rmse'],
'\nMAPE:\t', simulation1[ma]['final']['mae'])#,
# '\nMAPE:\t', simulation[ma]['final']['mape'])
# else:
# break
# code you want to evaluate
elapsed = timeit.default_timer() - start_time
print('Runtime: mins:',elapsed/60)
SMA
SMA([input_arrays], [timeperiod=30])
Simple Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
17
Working on SMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.76 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4157.020, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3687.148, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.28 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3458.651, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3322.133, Time=0.13 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=0.99 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.06 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3324.133, Time=0.28 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 3.706 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1657.067
Date: Sun, 12 Dec 2021 AIC 3322.133
Time: 11:55:23 BIC 3340.897
Sample: 0 HQIC 3329.339
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1966 0.003 -387.226 0.000 -1.203 -1.191
ar.L2 -0.8952 0.006 -138.692 0.000 -0.908 -0.883
ar.L3 -0.3968 0.006 -68.284 0.000 -0.408 -0.385
sigma2 3.5858 0.017 214.535 0.000 3.553 3.619
===================================================================================
Ljung-Box (L1) (Q): 14.47 Jarque-Bera (JB): 2428881.42
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 271.99
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
WARNING:tensorflow:Layer lstm will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04388, saving model to LSTM1.h5
48/48 - 4s - loss: 0.1044 - val_loss: 0.0439 - lr: 0.0010 - 4s/epoch - 86ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.04388
48/48 - 1s - loss: 0.0884 - val_loss: 0.1990 - lr: 0.0010 - 748ms/epoch - 16ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04388
48/48 - 1s - loss: 0.1522 - val_loss: 0.9226 - lr: 0.0010 - 703ms/epoch - 15ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04388
48/48 - 1s - loss: 0.0562 - val_loss: 0.2037 - lr: 0.0010 - 677ms/epoch - 14ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.04388 to 0.03243, saving model to LSTM1.h5
48/48 - 1s - loss: 0.0460 - val_loss: 0.0324 - lr: 0.0010 - 739ms/epoch - 15ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.03243
48/48 - 1s - loss: 0.0356 - val_loss: 0.0477 - lr: 0.0010 - 725ms/epoch - 15ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.03243 to 0.02299, saving model to LSTM1.h5
48/48 - 1s - loss: 0.0379 - val_loss: 0.0230 - lr: 0.0010 - 743ms/epoch - 15ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.02299 to 0.01352, saving model to LSTM1.h5
48/48 - 1s - loss: 0.0375 - val_loss: 0.0135 - lr: 0.0010 - 768ms/epoch - 16ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.01352
48/48 - 1s - loss: 0.0322 - val_loss: 0.0485 - lr: 0.0010 - 759ms/epoch - 16ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.01352
48/48 - 1s - loss: 0.0433 - val_loss: 0.1783 - lr: 0.0010 - 758ms/epoch - 16ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.01352
48/48 - 1s - loss: 0.0380 - val_loss: 0.3196 - lr: 0.0010 - 728ms/epoch - 15ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01352
48/48 - 1s - loss: 0.0362 - val_loss: 0.0764 - lr: 0.0010 - 749ms/epoch - 16ms/step
Epoch 13/500
Epoch 00013: val_loss improved from 0.01352 to 0.00713, saving model to LSTM1.h5
48/48 - 1s - loss: 0.0345 - val_loss: 0.0071 - lr: 0.0010 - 768ms/epoch - 16ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0355 - val_loss: 0.1058 - lr: 0.0010 - 705ms/epoch - 15ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0363 - val_loss: 0.0092 - lr: 0.0010 - 700ms/epoch - 15ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0346 - val_loss: 0.0292 - lr: 0.0010 - 700ms/epoch - 15ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0413 - val_loss: 0.0217 - lr: 0.0010 - 687ms/epoch - 14ms/step
Epoch 18/500
Epoch 00018: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00018: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0412 - val_loss: 0.0403 - lr: 0.0010 - 697ms/epoch - 15ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0490 - val_loss: 0.0240 - lr: 1.0000e-04 - 761ms/epoch - 16ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0303 - val_loss: 0.0284 - lr: 1.0000e-04 - 730ms/epoch - 15ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0264 - val_loss: 0.0344 - lr: 1.0000e-04 - 720ms/epoch - 15ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0239 - val_loss: 0.0400 - lr: 1.0000e-04 - 720ms/epoch - 15ms/step
Epoch 23/500
Epoch 00023: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00023: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0233 - val_loss: 0.0489 - lr: 1.0000e-04 - 792ms/epoch - 16ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0219 - val_loss: 0.0495 - lr: 1.0000e-05 - 776ms/epoch - 16ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0249 - val_loss: 0.0501 - lr: 1.0000e-05 - 764ms/epoch - 16ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0211 - val_loss: 0.0504 - lr: 1.0000e-05 - 748ms/epoch - 16ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0215 - val_loss: 0.0513 - lr: 1.0000e-05 - 738ms/epoch - 15ms/step
Epoch 28/500
Epoch 00028: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00028: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0230 - val_loss: 0.0509 - lr: 1.0000e-05 - 734ms/epoch - 15ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0214 - val_loss: 0.0511 - lr: 1.0000e-05 - 791ms/epoch - 16ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0247 - val_loss: 0.0505 - lr: 1.0000e-05 - 710ms/epoch - 15ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0227 - val_loss: 0.0517 - lr: 1.0000e-05 - 771ms/epoch - 16ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0240 - val_loss: 0.0515 - lr: 1.0000e-05 - 716ms/epoch - 15ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0233 - val_loss: 0.0518 - lr: 1.0000e-05 - 735ms/epoch - 15ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0235 - val_loss: 0.0508 - lr: 1.0000e-05 - 724ms/epoch - 15ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0243 - val_loss: 0.0520 - lr: 1.0000e-05 - 749ms/epoch - 16ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0223 - val_loss: 0.0546 - lr: 1.0000e-05 - 828ms/epoch - 17ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0226 - val_loss: 0.0559 - lr: 1.0000e-05 - 754ms/epoch - 16ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0236 - val_loss: 0.0569 - lr: 1.0000e-05 - 730ms/epoch - 15ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0228 - val_loss: 0.0577 - lr: 1.0000e-05 - 714ms/epoch - 15ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0216 - val_loss: 0.0567 - lr: 1.0000e-05 - 723ms/epoch - 15ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0269 - val_loss: 0.0589 - lr: 1.0000e-05 - 745ms/epoch - 16ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0227 - val_loss: 0.0590 - lr: 1.0000e-05 - 736ms/epoch - 15ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0234 - val_loss: 0.0602 - lr: 1.0000e-05 - 741ms/epoch - 15ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0239 - val_loss: 0.0624 - lr: 1.0000e-05 - 704ms/epoch - 15ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0233 - val_loss: 0.0619 - lr: 1.0000e-05 - 724ms/epoch - 15ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0218 - val_loss: 0.0605 - lr: 1.0000e-05 - 728ms/epoch - 15ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0206 - val_loss: 0.0594 - lr: 1.0000e-05 - 727ms/epoch - 15ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0232 - val_loss: 0.0592 - lr: 1.0000e-05 - 710ms/epoch - 15ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0215 - val_loss: 0.0577 - lr: 1.0000e-05 - 750ms/epoch - 16ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0207 - val_loss: 0.0575 - lr: 1.0000e-05 - 771ms/epoch - 16ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0221 - val_loss: 0.0570 - lr: 1.0000e-05 - 753ms/epoch - 16ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0240 - val_loss: 0.0578 - lr: 1.0000e-05 - 749ms/epoch - 16ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0229 - val_loss: 0.0587 - lr: 1.0000e-05 - 736ms/epoch - 15ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0226 - val_loss: 0.0599 - lr: 1.0000e-05 - 738ms/epoch - 15ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0226 - val_loss: 0.0606 - lr: 1.0000e-05 - 748ms/epoch - 16ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0213 - val_loss: 0.0609 - lr: 1.0000e-05 - 762ms/epoch - 16ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0212 - val_loss: 0.0603 - lr: 1.0000e-05 - 788ms/epoch - 16ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0233 - val_loss: 0.0585 - lr: 1.0000e-05 - 685ms/epoch - 14ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0217 - val_loss: 0.0596 - lr: 1.0000e-05 - 728ms/epoch - 15ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0234 - val_loss: 0.0609 - lr: 1.0000e-05 - 743ms/epoch - 15ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0222 - val_loss: 0.0614 - lr: 1.0000e-05 - 709ms/epoch - 15ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0231 - val_loss: 0.0612 - lr: 1.0000e-05 - 738ms/epoch - 15ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.00713
48/48 - 1s - loss: 0.0217 - val_loss: 0.0603 - lr: 1.0000e-05 - 737ms/epoch - 15ms/step
Epoch 00063: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 216.26215770312788
RMSE: 14.705854538350632
MAPE: 11.92367463073216
EMA
EMA([input_arrays], [timeperiod=30])
Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
51
Working on EMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.58 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4231.556, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3761.238, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.40 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3532.227, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3394.496, Time=0.15 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.14 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.89 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3396.496, Time=0.27 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 3.641 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1693.248
Date: Sun, 12 Dec 2021 AIC 3394.496
Time: 11:58:20 BIC 3413.260
Sample: 0 HQIC 3401.702
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1982 0.003 -389.569 0.000 -1.204 -1.192
ar.L2 -0.8976 0.006 -139.811 0.000 -0.910 -0.885
ar.L3 -0.3984 0.006 -68.662 0.000 -0.410 -0.387
sigma2 3.9230 0.018 215.372 0.000 3.887 3.959
===================================================================================
Ljung-Box (L1) (Q): 14.54 Jarque-Bera (JB): 2462173.05
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 273.82
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
WARNING:tensorflow:Layer lstm_1 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.78839, saving model to LSTM1.h5
16/16 - 2s - loss: 0.3784 - val_loss: 0.7884 - lr: 0.0010 - 2s/epoch - 140ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.78839 to 0.29586, saving model to LSTM1.h5
16/16 - 0s - loss: 0.0819 - val_loss: 0.2959 - lr: 0.0010 - 307ms/epoch - 19ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.29586 to 0.02176, saving model to LSTM1.h5
16/16 - 0s - loss: 0.0443 - val_loss: 0.0218 - lr: 0.0010 - 296ms/epoch - 19ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.02176 to 0.01234, saving model to LSTM1.h5
16/16 - 0s - loss: 0.0428 - val_loss: 0.0123 - lr: 0.0010 - 302ms/epoch - 19ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.01234 to 0.01154, saving model to LSTM1.h5
16/16 - 0s - loss: 0.0448 - val_loss: 0.0115 - lr: 0.0010 - 303ms/epoch - 19ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.01154 to 0.01000, saving model to LSTM1.h5
16/16 - 0s - loss: 0.0398 - val_loss: 0.0100 - lr: 0.0010 - 352ms/epoch - 22ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0341 - val_loss: 0.0144 - lr: 0.0010 - 271ms/epoch - 17ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0350 - val_loss: 0.0313 - lr: 0.0010 - 260ms/epoch - 16ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0453 - val_loss: 0.0103 - lr: 0.0010 - 269ms/epoch - 17ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0546 - val_loss: 0.0115 - lr: 0.0010 - 306ms/epoch - 19ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00011: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0359 - val_loss: 0.0135 - lr: 0.0010 - 281ms/epoch - 18ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0346 - val_loss: 0.0124 - lr: 1.0000e-04 - 256ms/epoch - 16ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0313 - val_loss: 0.0119 - lr: 1.0000e-04 - 299ms/epoch - 19ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0293 - val_loss: 0.0127 - lr: 1.0000e-04 - 270ms/epoch - 17ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0311 - val_loss: 0.0126 - lr: 1.0000e-04 - 264ms/epoch - 17ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00016: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0275 - val_loss: 0.0129 - lr: 1.0000e-04 - 315ms/epoch - 20ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0291 - val_loss: 0.0129 - lr: 1.0000e-05 - 264ms/epoch - 17ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0307 - val_loss: 0.0128 - lr: 1.0000e-05 - 283ms/epoch - 18ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0299 - val_loss: 0.0128 - lr: 1.0000e-05 - 269ms/epoch - 17ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0280 - val_loss: 0.0130 - lr: 1.0000e-05 - 266ms/epoch - 17ms/step
Epoch 21/500
Epoch 00021: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00021: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0283 - val_loss: 0.0131 - lr: 1.0000e-05 - 277ms/epoch - 17ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0271 - val_loss: 0.0131 - lr: 1.0000e-05 - 273ms/epoch - 17ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0278 - val_loss: 0.0133 - lr: 1.0000e-05 - 273ms/epoch - 17ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0314 - val_loss: 0.0134 - lr: 1.0000e-05 - 276ms/epoch - 17ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0307 - val_loss: 0.0136 - lr: 1.0000e-05 - 280ms/epoch - 17ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0287 - val_loss: 0.0136 - lr: 1.0000e-05 - 254ms/epoch - 16ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0284 - val_loss: 0.0137 - lr: 1.0000e-05 - 273ms/epoch - 17ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0259 - val_loss: 0.0137 - lr: 1.0000e-05 - 291ms/epoch - 18ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0275 - val_loss: 0.0136 - lr: 1.0000e-05 - 266ms/epoch - 17ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0279 - val_loss: 0.0135 - lr: 1.0000e-05 - 263ms/epoch - 16ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0286 - val_loss: 0.0135 - lr: 1.0000e-05 - 292ms/epoch - 18ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0275 - val_loss: 0.0134 - lr: 1.0000e-05 - 312ms/epoch - 19ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0287 - val_loss: 0.0132 - lr: 1.0000e-05 - 285ms/epoch - 18ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0307 - val_loss: 0.0133 - lr: 1.0000e-05 - 264ms/epoch - 16ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0294 - val_loss: 0.0135 - lr: 1.0000e-05 - 276ms/epoch - 17ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0297 - val_loss: 0.0137 - lr: 1.0000e-05 - 310ms/epoch - 19ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0267 - val_loss: 0.0138 - lr: 1.0000e-05 - 314ms/epoch - 20ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0296 - val_loss: 0.0139 - lr: 1.0000e-05 - 260ms/epoch - 16ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0292 - val_loss: 0.0137 - lr: 1.0000e-05 - 284ms/epoch - 18ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0284 - val_loss: 0.0136 - lr: 1.0000e-05 - 256ms/epoch - 16ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0279 - val_loss: 0.0134 - lr: 1.0000e-05 - 279ms/epoch - 17ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0285 - val_loss: 0.0136 - lr: 1.0000e-05 - 290ms/epoch - 18ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0307 - val_loss: 0.0137 - lr: 1.0000e-05 - 256ms/epoch - 16ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0285 - val_loss: 0.0138 - lr: 1.0000e-05 - 290ms/epoch - 18ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0260 - val_loss: 0.0139 - lr: 1.0000e-05 - 318ms/epoch - 20ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0296 - val_loss: 0.0136 - lr: 1.0000e-05 - 296ms/epoch - 19ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0284 - val_loss: 0.0136 - lr: 1.0000e-05 - 278ms/epoch - 17ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0293 - val_loss: 0.0137 - lr: 1.0000e-05 - 286ms/epoch - 18ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0265 - val_loss: 0.0138 - lr: 1.0000e-05 - 324ms/epoch - 20ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0277 - val_loss: 0.0141 - lr: 1.0000e-05 - 300ms/epoch - 19ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0279 - val_loss: 0.0142 - lr: 1.0000e-05 - 316ms/epoch - 20ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0289 - val_loss: 0.0142 - lr: 1.0000e-05 - 285ms/epoch - 18ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0290 - val_loss: 0.0143 - lr: 1.0000e-05 - 255ms/epoch - 16ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0272 - val_loss: 0.0144 - lr: 1.0000e-05 - 292ms/epoch - 18ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0271 - val_loss: 0.0146 - lr: 1.0000e-05 - 277ms/epoch - 17ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.01000
16/16 - 0s - loss: 0.0272 - val_loss: 0.0148 - lr: 1.0000e-05 - 291ms/epoch - 18ms/step
Epoch 00056: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 216.26215770312788
RMSE: 14.705854538350632
MAPE: 11.92367463073216
EMA
Prediction vs Close: 54.1% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 122.96033735804009
RMSE: 11.08874823224155
MAPE: 9.251696034357076
WMA
WMA([input_arrays], [timeperiod=30])
Weighted Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
49
Working on WMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.58 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4264.089, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3793.930, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.34 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3564.923, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3427.258, Time=0.12 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.69 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.61 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3429.258, Time=0.24 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 3.787 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1709.629
Date: Sun, 12 Dec 2021 AIC 3427.258
Time: 12:00:12 BIC 3446.021
Sample: 0 HQIC 3434.464
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1981 0.003 -389.386 0.000 -1.204 -1.192
ar.L2 -0.8974 0.006 -139.699 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.737 0.000 -0.410 -0.387
sigma2 4.0860 0.019 215.311 0.000 4.049 4.123
===================================================================================
Ljung-Box (L1) (Q): 14.57 Jarque-Bera (JB): 2460901.70
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 273.75
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
WARNING:tensorflow:Layer lstm_2 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.03569, saving model to LSTM1.h5
17/17 - 2s - loss: 0.2460 - val_loss: 0.0357 - lr: 0.0010 - 2s/epoch - 133ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.03569
17/17 - 0s - loss: 0.1094 - val_loss: 1.4134 - lr: 0.0010 - 303ms/epoch - 18ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.03569 to 0.03486, saving model to LSTM1.h5
17/17 - 0s - loss: 0.0632 - val_loss: 0.0349 - lr: 0.0010 - 310ms/epoch - 18ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.03486
17/17 - 0s - loss: 0.0723 - val_loss: 0.0602 - lr: 0.0010 - 286ms/epoch - 17ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.03486 to 0.01947, saving model to LSTM1.h5
17/17 - 0s - loss: 0.0427 - val_loss: 0.0195 - lr: 0.0010 - 318ms/epoch - 19ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.01947 to 0.00806, saving model to LSTM1.h5
17/17 - 0s - loss: 0.0412 - val_loss: 0.0081 - lr: 0.0010 - 301ms/epoch - 18ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0373 - val_loss: 0.0109 - lr: 0.0010 - 310ms/epoch - 18ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0426 - val_loss: 0.0607 - lr: 0.0010 - 312ms/epoch - 18ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0362 - val_loss: 0.0176 - lr: 0.0010 - 291ms/epoch - 17ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0343 - val_loss: 0.0269 - lr: 0.0010 - 312ms/epoch - 18ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00011: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0343 - val_loss: 0.0276 - lr: 0.0010 - 278ms/epoch - 16ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0320 - val_loss: 0.0273 - lr: 1.0000e-04 - 290ms/epoch - 17ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0300 - val_loss: 0.0293 - lr: 1.0000e-04 - 287ms/epoch - 17ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0323 - val_loss: 0.0290 - lr: 1.0000e-04 - 279ms/epoch - 16ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0318 - val_loss: 0.0291 - lr: 1.0000e-04 - 305ms/epoch - 18ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00016: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0302 - val_loss: 0.0296 - lr: 1.0000e-04 - 281ms/epoch - 17ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0328 - val_loss: 0.0294 - lr: 1.0000e-05 - 305ms/epoch - 18ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0314 - val_loss: 0.0292 - lr: 1.0000e-05 - 263ms/epoch - 15ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0324 - val_loss: 0.0290 - lr: 1.0000e-05 - 279ms/epoch - 16ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0311 - val_loss: 0.0287 - lr: 1.0000e-05 - 293ms/epoch - 17ms/step
Epoch 21/500
Epoch 00021: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00021: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0272 - val_loss: 0.0284 - lr: 1.0000e-05 - 278ms/epoch - 16ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0348 - val_loss: 0.0283 - lr: 1.0000e-05 - 293ms/epoch - 17ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0306 - val_loss: 0.0282 - lr: 1.0000e-05 - 279ms/epoch - 16ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0301 - val_loss: 0.0281 - lr: 1.0000e-05 - 287ms/epoch - 17ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0304 - val_loss: 0.0281 - lr: 1.0000e-05 - 301ms/epoch - 18ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0313 - val_loss: 0.0280 - lr: 1.0000e-05 - 292ms/epoch - 17ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0299 - val_loss: 0.0280 - lr: 1.0000e-05 - 290ms/epoch - 17ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0331 - val_loss: 0.0283 - lr: 1.0000e-05 - 286ms/epoch - 17ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0322 - val_loss: 0.0283 - lr: 1.0000e-05 - 299ms/epoch - 18ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0309 - val_loss: 0.0283 - lr: 1.0000e-05 - 300ms/epoch - 18ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0330 - val_loss: 0.0280 - lr: 1.0000e-05 - 290ms/epoch - 17ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0301 - val_loss: 0.0278 - lr: 1.0000e-05 - 277ms/epoch - 16ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0301 - val_loss: 0.0279 - lr: 1.0000e-05 - 283ms/epoch - 17ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0314 - val_loss: 0.0278 - lr: 1.0000e-05 - 315ms/epoch - 19ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0322 - val_loss: 0.0275 - lr: 1.0000e-05 - 294ms/epoch - 17ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0306 - val_loss: 0.0277 - lr: 1.0000e-05 - 284ms/epoch - 17ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0310 - val_loss: 0.0279 - lr: 1.0000e-05 - 282ms/epoch - 17ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0331 - val_loss: 0.0278 - lr: 1.0000e-05 - 316ms/epoch - 19ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0289 - val_loss: 0.0275 - lr: 1.0000e-05 - 310ms/epoch - 18ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0318 - val_loss: 0.0274 - lr: 1.0000e-05 - 292ms/epoch - 17ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0336 - val_loss: 0.0272 - lr: 1.0000e-05 - 276ms/epoch - 16ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0314 - val_loss: 0.0271 - lr: 1.0000e-05 - 267ms/epoch - 16ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0332 - val_loss: 0.0273 - lr: 1.0000e-05 - 297ms/epoch - 17ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0308 - val_loss: 0.0274 - lr: 1.0000e-05 - 267ms/epoch - 16ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0286 - val_loss: 0.0276 - lr: 1.0000e-05 - 277ms/epoch - 16ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0331 - val_loss: 0.0276 - lr: 1.0000e-05 - 289ms/epoch - 17ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0338 - val_loss: 0.0276 - lr: 1.0000e-05 - 279ms/epoch - 16ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0327 - val_loss: 0.0279 - lr: 1.0000e-05 - 287ms/epoch - 17ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0331 - val_loss: 0.0286 - lr: 1.0000e-05 - 268ms/epoch - 16ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0341 - val_loss: 0.0287 - lr: 1.0000e-05 - 300ms/epoch - 18ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0292 - val_loss: 0.0288 - lr: 1.0000e-05 - 290ms/epoch - 17ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0316 - val_loss: 0.0293 - lr: 1.0000e-05 - 283ms/epoch - 17ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0341 - val_loss: 0.0292 - lr: 1.0000e-05 - 290ms/epoch - 17ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0284 - val_loss: 0.0290 - lr: 1.0000e-05 - 293ms/epoch - 17ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0318 - val_loss: 0.0287 - lr: 1.0000e-05 - 267ms/epoch - 16ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00806
17/17 - 0s - loss: 0.0329 - val_loss: 0.0286 - lr: 1.0000e-05 - 299ms/epoch - 18ms/step
Epoch 00056: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 216.26215770312788
RMSE: 14.705854538350632
MAPE: 11.92367463073216
EMA
Prediction vs Close: 54.1% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 122.96033735804009
RMSE: 11.08874823224155
MAPE: 9.251696034357076
WMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 56.80196950717294
RMSE: 7.5367081346681415
MAPE: 5.956122993340066
DEMA
DEMA([input_arrays], [timeperiod=30])
Double Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
89
Working on DEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.55 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4436.126, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3965.317, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.53 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3736.589, Time=0.09 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3598.951, Time=0.11 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.27 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.26 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3600.951, Time=0.60 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.515 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1795.475
Date: Sun, 12 Dec 2021 AIC 3598.951
Time: 12:02:04 BIC 3617.714
Sample: 0 HQIC 3606.157
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1983 0.003 -389.581 0.000 -1.204 -1.192
ar.L2 -0.8973 0.006 -139.732 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.649 0.000 -0.410 -0.387
sigma2 5.0573 0.023 215.292 0.000 5.011 5.103
===================================================================================
Ljung-Box (L1) (Q): 14.41 Jarque-Bera (JB): 2460553.80
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.89
Prob(H) (two-sided): 0.00 Kurtosis: 273.74
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
WARNING:tensorflow:Layer lstm_3 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.24906, saving model to LSTM1.h5
10/10 - 2s - loss: 0.4063 - val_loss: 0.2491 - lr: 0.0010 - 2s/epoch - 211ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.24906 to 0.02406, saving model to LSTM1.h5
10/10 - 0s - loss: 0.1642 - val_loss: 0.0241 - lr: 0.0010 - 228ms/epoch - 23ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.02406
10/10 - 0s - loss: 0.0864 - val_loss: 0.3861 - lr: 0.0010 - 175ms/epoch - 18ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.02406
10/10 - 0s - loss: 0.0632 - val_loss: 0.1415 - lr: 0.0010 - 183ms/epoch - 18ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.02406 to 0.01876, saving model to LSTM1.h5
10/10 - 0s - loss: 0.0590 - val_loss: 0.0188 - lr: 0.0010 - 225ms/epoch - 23ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0546 - val_loss: 0.0764 - lr: 0.0010 - 187ms/epoch - 19ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0422 - val_loss: 0.1088 - lr: 0.0010 - 192ms/epoch - 19ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0399 - val_loss: 0.0605 - lr: 0.0010 - 186ms/epoch - 19ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0353 - val_loss: 0.0408 - lr: 0.0010 - 192ms/epoch - 19ms/step
Epoch 10/500
Epoch 00010: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00010: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0326 - val_loss: 0.0544 - lr: 0.0010 - 188ms/epoch - 19ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0312 - val_loss: 0.0509 - lr: 1.0000e-04 - 181ms/epoch - 18ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0296 - val_loss: 0.0496 - lr: 1.0000e-04 - 190ms/epoch - 19ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0280 - val_loss: 0.0484 - lr: 1.0000e-04 - 182ms/epoch - 18ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0287 - val_loss: 0.0494 - lr: 1.0000e-04 - 189ms/epoch - 19ms/step
Epoch 15/500
Epoch 00015: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00015: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0273 - val_loss: 0.0539 - lr: 1.0000e-04 - 193ms/epoch - 19ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0292 - val_loss: 0.0539 - lr: 1.0000e-05 - 196ms/epoch - 20ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0278 - val_loss: 0.0535 - lr: 1.0000e-05 - 194ms/epoch - 19ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0277 - val_loss: 0.0530 - lr: 1.0000e-05 - 173ms/epoch - 17ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0298 - val_loss: 0.0525 - lr: 1.0000e-05 - 193ms/epoch - 19ms/step
Epoch 20/500
Epoch 00020: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00020: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0287 - val_loss: 0.0522 - lr: 1.0000e-05 - 242ms/epoch - 24ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0304 - val_loss: 0.0520 - lr: 1.0000e-05 - 185ms/epoch - 19ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0275 - val_loss: 0.0515 - lr: 1.0000e-05 - 187ms/epoch - 19ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0279 - val_loss: 0.0513 - lr: 1.0000e-05 - 170ms/epoch - 17ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0314 - val_loss: 0.0509 - lr: 1.0000e-05 - 183ms/epoch - 18ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0275 - val_loss: 0.0503 - lr: 1.0000e-05 - 190ms/epoch - 19ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0280 - val_loss: 0.0496 - lr: 1.0000e-05 - 198ms/epoch - 20ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0281 - val_loss: 0.0488 - lr: 1.0000e-05 - 166ms/epoch - 17ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0293 - val_loss: 0.0485 - lr: 1.0000e-05 - 192ms/epoch - 19ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0305 - val_loss: 0.0480 - lr: 1.0000e-05 - 192ms/epoch - 19ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0283 - val_loss: 0.0475 - lr: 1.0000e-05 - 193ms/epoch - 19ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0289 - val_loss: 0.0473 - lr: 1.0000e-05 - 203ms/epoch - 20ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0307 - val_loss: 0.0473 - lr: 1.0000e-05 - 190ms/epoch - 19ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0314 - val_loss: 0.0473 - lr: 1.0000e-05 - 193ms/epoch - 19ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0279 - val_loss: 0.0468 - lr: 1.0000e-05 - 186ms/epoch - 19ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0268 - val_loss: 0.0467 - lr: 1.0000e-05 - 177ms/epoch - 18ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0255 - val_loss: 0.0469 - lr: 1.0000e-05 - 195ms/epoch - 20ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0294 - val_loss: 0.0467 - lr: 1.0000e-05 - 174ms/epoch - 17ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0271 - val_loss: 0.0463 - lr: 1.0000e-05 - 178ms/epoch - 18ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0292 - val_loss: 0.0461 - lr: 1.0000e-05 - 181ms/epoch - 18ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0291 - val_loss: 0.0456 - lr: 1.0000e-05 - 196ms/epoch - 20ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0291 - val_loss: 0.0452 - lr: 1.0000e-05 - 220ms/epoch - 22ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0270 - val_loss: 0.0447 - lr: 1.0000e-05 - 187ms/epoch - 19ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0285 - val_loss: 0.0443 - lr: 1.0000e-05 - 180ms/epoch - 18ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0288 - val_loss: 0.0441 - lr: 1.0000e-05 - 194ms/epoch - 19ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0273 - val_loss: 0.0434 - lr: 1.0000e-05 - 186ms/epoch - 19ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0309 - val_loss: 0.0426 - lr: 1.0000e-05 - 193ms/epoch - 19ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0326 - val_loss: 0.0418 - lr: 1.0000e-05 - 197ms/epoch - 20ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0284 - val_loss: 0.0415 - lr: 1.0000e-05 - 175ms/epoch - 18ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0297 - val_loss: 0.0420 - lr: 1.0000e-05 - 186ms/epoch - 19ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0293 - val_loss: 0.0421 - lr: 1.0000e-05 - 182ms/epoch - 18ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0275 - val_loss: 0.0419 - lr: 1.0000e-05 - 195ms/epoch - 19ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0271 - val_loss: 0.0418 - lr: 1.0000e-05 - 190ms/epoch - 19ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0310 - val_loss: 0.0416 - lr: 1.0000e-05 - 182ms/epoch - 18ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0271 - val_loss: 0.0417 - lr: 1.0000e-05 - 179ms/epoch - 18ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.01876
10/10 - 0s - loss: 0.0273 - val_loss: 0.0416 - lr: 1.0000e-05 - 185ms/epoch - 19ms/step
Epoch 00055: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 216.26215770312788
RMSE: 14.705854538350632
MAPE: 11.92367463073216
EMA
Prediction vs Close: 54.1% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 122.96033735804009
RMSE: 11.08874823224155
MAPE: 9.251696034357076
WMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 56.80196950717294
RMSE: 7.5367081346681415
MAPE: 5.956122993340066
DEMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 42.16% Accuracy
MSE: 120.51656357671082
RMSE: 10.978003624371365
MAPE: 9.343426819843298
KAMA
KAMA([input_arrays], [timeperiod=30])
Kaufman Adaptive Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
18
Working on KAMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.50 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4190.464, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3724.371, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.38 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3494.154, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3357.435, Time=0.13 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.62 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.99 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3359.435, Time=0.27 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.093 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1674.717
Date: Sun, 12 Dec 2021 AIC 3357.435
Time: 12:03:48 BIC 3376.198
Sample: 0 HQIC 3364.641
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1955 0.003 -381.246 0.000 -1.202 -1.189
ar.L2 -0.8964 0.007 -135.835 0.000 -0.909 -0.883
ar.L3 -0.3971 0.006 -67.229 0.000 -0.409 -0.385
sigma2 3.7466 0.018 211.623 0.000 3.712 3.781
===================================================================================
Ljung-Box (L1) (Q): 14.20 Jarque-Bera (JB): 2338363.32
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.01 Skew: 3.76
Prob(H) (two-sided): 0.00 Kurtosis: 266.93
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
WARNING:tensorflow:Layer lstm_4 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.06791, saving model to LSTM1.h5
45/45 - 3s - loss: 0.4313 - val_loss: 0.0679 - lr: 0.0010 - 3s/epoch - 60ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.06791
45/45 - 1s - loss: 0.0693 - val_loss: 0.1067 - lr: 0.0010 - 714ms/epoch - 16ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.06791
45/45 - 1s - loss: 0.0562 - val_loss: 0.1727 - lr: 0.0010 - 802ms/epoch - 18ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.06791
45/45 - 1s - loss: 0.0732 - val_loss: 0.7143 - lr: 0.0010 - 751ms/epoch - 17ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.06791
45/45 - 1s - loss: 0.0487 - val_loss: 0.3365 - lr: 0.0010 - 725ms/epoch - 16ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.06791
45/45 - 1s - loss: 0.0416 - val_loss: 0.1184 - lr: 0.0010 - 734ms/epoch - 16ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.06791
45/45 - 1s - loss: 0.0379 - val_loss: 0.0948 - lr: 1.0000e-04 - 721ms/epoch - 16ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.06791
45/45 - 1s - loss: 0.0372 - val_loss: 0.0775 - lr: 1.0000e-04 - 730ms/epoch - 16ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.06791 to 0.06647, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0373 - val_loss: 0.0665 - lr: 1.0000e-04 - 765ms/epoch - 17ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.06647 to 0.05375, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0354 - val_loss: 0.0537 - lr: 1.0000e-04 - 752ms/epoch - 17ms/step
Epoch 11/500
Epoch 00011: val_loss improved from 0.05375 to 0.04523, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0356 - val_loss: 0.0452 - lr: 1.0000e-04 - 765ms/epoch - 17ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.04523
45/45 - 1s - loss: 0.0363 - val_loss: 0.0467 - lr: 1.0000e-04 - 698ms/epoch - 16ms/step
Epoch 13/500
Epoch 00013: val_loss improved from 0.04523 to 0.04262, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0316 - val_loss: 0.0426 - lr: 1.0000e-04 - 799ms/epoch - 18ms/step
Epoch 14/500
Epoch 00014: val_loss improved from 0.04262 to 0.03912, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0312 - val_loss: 0.0391 - lr: 1.0000e-04 - 716ms/epoch - 16ms/step
Epoch 15/500
Epoch 00015: val_loss improved from 0.03912 to 0.03484, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0303 - val_loss: 0.0348 - lr: 1.0000e-04 - 734ms/epoch - 16ms/step
Epoch 16/500
Epoch 00016: val_loss improved from 0.03484 to 0.02902, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0328 - val_loss: 0.0290 - lr: 1.0000e-04 - 741ms/epoch - 16ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.02902
45/45 - 1s - loss: 0.0356 - val_loss: 0.0350 - lr: 1.0000e-04 - 727ms/epoch - 16ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.02902
45/45 - 1s - loss: 0.0318 - val_loss: 0.0321 - lr: 1.0000e-04 - 731ms/epoch - 16ms/step
Epoch 19/500
Epoch 00019: val_loss improved from 0.02902 to 0.02669, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0292 - val_loss: 0.0267 - lr: 1.0000e-04 - 732ms/epoch - 16ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.02669
45/45 - 1s - loss: 0.0297 - val_loss: 0.0281 - lr: 1.0000e-04 - 763ms/epoch - 17ms/step
Epoch 21/500
Epoch 00021: val_loss improved from 0.02669 to 0.02659, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0306 - val_loss: 0.0266 - lr: 1.0000e-04 - 766ms/epoch - 17ms/step
Epoch 22/500
Epoch 00022: val_loss improved from 0.02659 to 0.02457, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0284 - val_loss: 0.0246 - lr: 1.0000e-04 - 751ms/epoch - 17ms/step
Epoch 23/500
Epoch 00023: val_loss improved from 0.02457 to 0.02332, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0299 - val_loss: 0.0233 - lr: 1.0000e-04 - 738ms/epoch - 16ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0298 - val_loss: 0.0303 - lr: 1.0000e-04 - 721ms/epoch - 16ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0289 - val_loss: 0.0270 - lr: 1.0000e-04 - 695ms/epoch - 15ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0292 - val_loss: 0.0320 - lr: 1.0000e-04 - 785ms/epoch - 17ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0272 - val_loss: 0.0283 - lr: 1.0000e-04 - 723ms/epoch - 16ms/step
Epoch 28/500
Epoch 00028: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00028: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0312 - val_loss: 0.0236 - lr: 1.0000e-04 - 715ms/epoch - 16ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0306 - val_loss: 0.0234 - lr: 1.0000e-05 - 742ms/epoch - 16ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0285 - val_loss: 0.0245 - lr: 1.0000e-05 - 761ms/epoch - 17ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0323 - val_loss: 0.0239 - lr: 1.0000e-05 - 728ms/epoch - 16ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0290 - val_loss: 0.0244 - lr: 1.0000e-05 - 735ms/epoch - 16ms/step
Epoch 33/500
Epoch 00033: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00033: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0275 - val_loss: 0.0250 - lr: 1.0000e-05 - 736ms/epoch - 16ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0266 - val_loss: 0.0247 - lr: 1.0000e-05 - 744ms/epoch - 17ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0279 - val_loss: 0.0250 - lr: 1.0000e-05 - 734ms/epoch - 16ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0276 - val_loss: 0.0249 - lr: 1.0000e-05 - 713ms/epoch - 16ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0287 - val_loss: 0.0254 - lr: 1.0000e-05 - 763ms/epoch - 17ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0275 - val_loss: 0.0260 - lr: 1.0000e-05 - 717ms/epoch - 16ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0295 - val_loss: 0.0267 - lr: 1.0000e-05 - 690ms/epoch - 15ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0313 - val_loss: 0.0269 - lr: 1.0000e-05 - 787ms/epoch - 17ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0277 - val_loss: 0.0268 - lr: 1.0000e-05 - 760ms/epoch - 17ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0283 - val_loss: 0.0270 - lr: 1.0000e-05 - 739ms/epoch - 16ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0272 - val_loss: 0.0270 - lr: 1.0000e-05 - 766ms/epoch - 17ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0295 - val_loss: 0.0264 - lr: 1.0000e-05 - 740ms/epoch - 16ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0296 - val_loss: 0.0263 - lr: 1.0000e-05 - 698ms/epoch - 16ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0295 - val_loss: 0.0267 - lr: 1.0000e-05 - 694ms/epoch - 15ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0280 - val_loss: 0.0263 - lr: 1.0000e-05 - 727ms/epoch - 16ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0262 - val_loss: 0.0259 - lr: 1.0000e-05 - 732ms/epoch - 16ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0286 - val_loss: 0.0265 - lr: 1.0000e-05 - 715ms/epoch - 16ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0270 - val_loss: 0.0265 - lr: 1.0000e-05 - 715ms/epoch - 16ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0273 - val_loss: 0.0258 - lr: 1.0000e-05 - 754ms/epoch - 17ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0262 - val_loss: 0.0251 - lr: 1.0000e-05 - 771ms/epoch - 17ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0297 - val_loss: 0.0242 - lr: 1.0000e-05 - 728ms/epoch - 16ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0287 - val_loss: 0.0238 - lr: 1.0000e-05 - 765ms/epoch - 17ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.02332
45/45 - 1s - loss: 0.0267 - val_loss: 0.0235 - lr: 1.0000e-05 - 751ms/epoch - 17ms/step
Epoch 56/500
Epoch 00056: val_loss improved from 0.02332 to 0.02322, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0265 - val_loss: 0.0232 - lr: 1.0000e-05 - 912ms/epoch - 20ms/step
Epoch 57/500
Epoch 00057: val_loss improved from 0.02322 to 0.02273, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0281 - val_loss: 0.0227 - lr: 1.0000e-05 - 793ms/epoch - 18ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.02273
45/45 - 1s - loss: 0.0288 - val_loss: 0.0228 - lr: 1.0000e-05 - 737ms/epoch - 16ms/step
Epoch 59/500
Epoch 00059: val_loss improved from 0.02273 to 0.02230, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0266 - val_loss: 0.0223 - lr: 1.0000e-05 - 704ms/epoch - 16ms/step
Epoch 60/500
Epoch 00060: val_loss improved from 0.02230 to 0.02186, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0273 - val_loss: 0.0219 - lr: 1.0000e-05 - 783ms/epoch - 17ms/step
Epoch 61/500
Epoch 00061: val_loss improved from 0.02186 to 0.02090, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0288 - val_loss: 0.0209 - lr: 1.0000e-05 - 787ms/epoch - 17ms/step
Epoch 62/500
Epoch 00062: val_loss improved from 0.02090 to 0.02068, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0285 - val_loss: 0.0207 - lr: 1.0000e-05 - 754ms/epoch - 17ms/step
Epoch 63/500
Epoch 00063: val_loss improved from 0.02068 to 0.02009, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0313 - val_loss: 0.0201 - lr: 1.0000e-05 - 774ms/epoch - 17ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.02009
45/45 - 1s - loss: 0.0274 - val_loss: 0.0203 - lr: 1.0000e-05 - 731ms/epoch - 16ms/step
Epoch 65/500
Epoch 00065: val_loss improved from 0.02009 to 0.01973, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0287 - val_loss: 0.0197 - lr: 1.0000e-05 - 736ms/epoch - 16ms/step
Epoch 66/500
Epoch 00066: val_loss improved from 0.01973 to 0.01902, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0277 - val_loss: 0.0190 - lr: 1.0000e-05 - 749ms/epoch - 17ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0298 - val_loss: 0.0192 - lr: 1.0000e-05 - 675ms/epoch - 15ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0270 - val_loss: 0.0191 - lr: 1.0000e-05 - 708ms/epoch - 16ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0272 - val_loss: 0.0196 - lr: 1.0000e-05 - 739ms/epoch - 16ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0317 - val_loss: 0.0202 - lr: 1.0000e-05 - 721ms/epoch - 16ms/step
Epoch 71/500
Epoch 00071: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0282 - val_loss: 0.0206 - lr: 1.0000e-05 - 743ms/epoch - 17ms/step
Epoch 72/500
Epoch 00072: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0279 - val_loss: 0.0205 - lr: 1.0000e-05 - 756ms/epoch - 17ms/step
Epoch 73/500
Epoch 00073: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0267 - val_loss: 0.0210 - lr: 1.0000e-05 - 707ms/epoch - 16ms/step
Epoch 74/500
Epoch 00074: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0256 - val_loss: 0.0210 - lr: 1.0000e-05 - 774ms/epoch - 17ms/step
Epoch 75/500
Epoch 00075: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0258 - val_loss: 0.0217 - lr: 1.0000e-05 - 718ms/epoch - 16ms/step
Epoch 76/500
Epoch 00076: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0255 - val_loss: 0.0212 - lr: 1.0000e-05 - 738ms/epoch - 16ms/step
Epoch 77/500
Epoch 00077: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0272 - val_loss: 0.0206 - lr: 1.0000e-05 - 722ms/epoch - 16ms/step
Epoch 78/500
Epoch 00078: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0263 - val_loss: 0.0205 - lr: 1.0000e-05 - 766ms/epoch - 17ms/step
Epoch 79/500
Epoch 00079: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0276 - val_loss: 0.0200 - lr: 1.0000e-05 - 714ms/epoch - 16ms/step
Epoch 80/500
Epoch 00080: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0281 - val_loss: 0.0198 - lr: 1.0000e-05 - 775ms/epoch - 17ms/step
Epoch 81/500
Epoch 00081: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0280 - val_loss: 0.0196 - lr: 1.0000e-05 - 741ms/epoch - 16ms/step
Epoch 82/500
Epoch 00082: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0287 - val_loss: 0.0192 - lr: 1.0000e-05 - 775ms/epoch - 17ms/step
Epoch 83/500
Epoch 00083: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0251 - val_loss: 0.0196 - lr: 1.0000e-05 - 739ms/epoch - 16ms/step
Epoch 84/500
Epoch 00084: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0280 - val_loss: 0.0196 - lr: 1.0000e-05 - 715ms/epoch - 16ms/step
Epoch 85/500
Epoch 00085: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0267 - val_loss: 0.0204 - lr: 1.0000e-05 - 729ms/epoch - 16ms/step
Epoch 86/500
Epoch 00086: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0278 - val_loss: 0.0205 - lr: 1.0000e-05 - 710ms/epoch - 16ms/step
Epoch 87/500
Epoch 00087: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0263 - val_loss: 0.0209 - lr: 1.0000e-05 - 721ms/epoch - 16ms/step
Epoch 88/500
Epoch 00088: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0264 - val_loss: 0.0209 - lr: 1.0000e-05 - 716ms/epoch - 16ms/step
Epoch 89/500
Epoch 00089: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0272 - val_loss: 0.0200 - lr: 1.0000e-05 - 710ms/epoch - 16ms/step
Epoch 90/500
Epoch 00090: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0275 - val_loss: 0.0197 - lr: 1.0000e-05 - 731ms/epoch - 16ms/step
Epoch 91/500
Epoch 00091: val_loss did not improve from 0.01902
45/45 - 1s - loss: 0.0245 - val_loss: 0.0201 - lr: 1.0000e-05 - 764ms/epoch - 17ms/step
Epoch 92/500
Epoch 00092: val_loss improved from 0.01902 to 0.01898, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0278 - val_loss: 0.0190 - lr: 1.0000e-05 - 931ms/epoch - 21ms/step
Epoch 93/500
Epoch 00093: val_loss improved from 0.01898 to 0.01893, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0284 - val_loss: 0.0189 - lr: 1.0000e-05 - 752ms/epoch - 17ms/step
Epoch 94/500
Epoch 00094: val_loss did not improve from 0.01893
45/45 - 1s - loss: 0.0278 - val_loss: 0.0197 - lr: 1.0000e-05 - 718ms/epoch - 16ms/step
Epoch 95/500
Epoch 00095: val_loss did not improve from 0.01893
45/45 - 1s - loss: 0.0268 - val_loss: 0.0191 - lr: 1.0000e-05 - 713ms/epoch - 16ms/step
Epoch 96/500
Epoch 00096: val_loss did not improve from 0.01893
45/45 - 1s - loss: 0.0274 - val_loss: 0.0197 - lr: 1.0000e-05 - 721ms/epoch - 16ms/step
Epoch 97/500
Epoch 00097: val_loss improved from 0.01893 to 0.01869, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0289 - val_loss: 0.0187 - lr: 1.0000e-05 - 744ms/epoch - 17ms/step
Epoch 98/500
Epoch 00098: val_loss improved from 0.01869 to 0.01787, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0242 - val_loss: 0.0179 - lr: 1.0000e-05 - 715ms/epoch - 16ms/step
Epoch 99/500
Epoch 00099: val_loss improved from 0.01787 to 0.01691, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0270 - val_loss: 0.0169 - lr: 1.0000e-05 - 719ms/epoch - 16ms/step
Epoch 100/500
Epoch 00100: val_loss improved from 0.01691 to 0.01676, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0277 - val_loss: 0.0168 - lr: 1.0000e-05 - 752ms/epoch - 17ms/step
Epoch 101/500
Epoch 00101: val_loss improved from 0.01676 to 0.01640, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0262 - val_loss: 0.0164 - lr: 1.0000e-05 - 751ms/epoch - 17ms/step
Epoch 102/500
Epoch 00102: val_loss improved from 0.01640 to 0.01636, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0288 - val_loss: 0.0164 - lr: 1.0000e-05 - 747ms/epoch - 17ms/step
Epoch 103/500
Epoch 00103: val_loss improved from 0.01636 to 0.01602, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0272 - val_loss: 0.0160 - lr: 1.0000e-05 - 691ms/epoch - 15ms/step
Epoch 104/500
Epoch 00104: val_loss did not improve from 0.01602
45/45 - 1s - loss: 0.0269 - val_loss: 0.0170 - lr: 1.0000e-05 - 710ms/epoch - 16ms/step
Epoch 105/500
Epoch 00105: val_loss did not improve from 0.01602
45/45 - 1s - loss: 0.0253 - val_loss: 0.0167 - lr: 1.0000e-05 - 725ms/epoch - 16ms/step
Epoch 106/500
Epoch 00106: val_loss did not improve from 0.01602
45/45 - 1s - loss: 0.0247 - val_loss: 0.0163 - lr: 1.0000e-05 - 718ms/epoch - 16ms/step
Epoch 107/500
Epoch 00107: val_loss improved from 0.01602 to 0.01594, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0257 - val_loss: 0.0159 - lr: 1.0000e-05 - 739ms/epoch - 16ms/step
Epoch 108/500
Epoch 00108: val_loss improved from 0.01594 to 0.01534, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0254 - val_loss: 0.0153 - lr: 1.0000e-05 - 767ms/epoch - 17ms/step
Epoch 109/500
Epoch 00109: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0254 - val_loss: 0.0161 - lr: 1.0000e-05 - 693ms/epoch - 15ms/step
Epoch 110/500
Epoch 00110: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0256 - val_loss: 0.0162 - lr: 1.0000e-05 - 694ms/epoch - 15ms/step
Epoch 111/500
Epoch 00111: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0255 - val_loss: 0.0170 - lr: 1.0000e-05 - 723ms/epoch - 16ms/step
Epoch 112/500
Epoch 00112: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0282 - val_loss: 0.0180 - lr: 1.0000e-05 - 701ms/epoch - 16ms/step
Epoch 113/500
Epoch 00113: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0281 - val_loss: 0.0180 - lr: 1.0000e-05 - 723ms/epoch - 16ms/step
Epoch 114/500
Epoch 00114: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0262 - val_loss: 0.0185 - lr: 1.0000e-05 - 730ms/epoch - 16ms/step
Epoch 115/500
Epoch 00115: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0278 - val_loss: 0.0180 - lr: 1.0000e-05 - 726ms/epoch - 16ms/step
Epoch 116/500
Epoch 00116: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0268 - val_loss: 0.0185 - lr: 1.0000e-05 - 752ms/epoch - 17ms/step
Epoch 117/500
Epoch 00117: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0276 - val_loss: 0.0180 - lr: 1.0000e-05 - 740ms/epoch - 16ms/step
Epoch 118/500
Epoch 00118: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0245 - val_loss: 0.0177 - lr: 1.0000e-05 - 756ms/epoch - 17ms/step
Epoch 119/500
Epoch 00119: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0255 - val_loss: 0.0171 - lr: 1.0000e-05 - 719ms/epoch - 16ms/step
Epoch 120/500
Epoch 00120: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0277 - val_loss: 0.0169 - lr: 1.0000e-05 - 719ms/epoch - 16ms/step
Epoch 121/500
Epoch 00121: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0252 - val_loss: 0.0172 - lr: 1.0000e-05 - 691ms/epoch - 15ms/step
Epoch 122/500
Epoch 00122: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0271 - val_loss: 0.0175 - lr: 1.0000e-05 - 722ms/epoch - 16ms/step
Epoch 123/500
Epoch 00123: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0281 - val_loss: 0.0170 - lr: 1.0000e-05 - 722ms/epoch - 16ms/step
Epoch 124/500
Epoch 00124: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0267 - val_loss: 0.0169 - lr: 1.0000e-05 - 733ms/epoch - 16ms/step
Epoch 125/500
Epoch 00125: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0245 - val_loss: 0.0156 - lr: 1.0000e-05 - 699ms/epoch - 16ms/step
Epoch 126/500
Epoch 00126: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0258 - val_loss: 0.0156 - lr: 1.0000e-05 - 761ms/epoch - 17ms/step
Epoch 127/500
Epoch 00127: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0271 - val_loss: 0.0156 - lr: 1.0000e-05 - 720ms/epoch - 16ms/step
Epoch 128/500
Epoch 00128: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0276 - val_loss: 0.0161 - lr: 1.0000e-05 - 786ms/epoch - 17ms/step
Epoch 129/500
Epoch 00129: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0277 - val_loss: 0.0155 - lr: 1.0000e-05 - 743ms/epoch - 17ms/step
Epoch 130/500
Epoch 00130: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0243 - val_loss: 0.0160 - lr: 1.0000e-05 - 717ms/epoch - 16ms/step
Epoch 131/500
Epoch 00131: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0263 - val_loss: 0.0165 - lr: 1.0000e-05 - 731ms/epoch - 16ms/step
Epoch 132/500
Epoch 00132: val_loss did not improve from 0.01534
45/45 - 1s - loss: 0.0244 - val_loss: 0.0160 - lr: 1.0000e-05 - 730ms/epoch - 16ms/step
Epoch 133/500
Epoch 00133: val_loss improved from 0.01534 to 0.01410, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0230 - val_loss: 0.0141 - lr: 1.0000e-05 - 944ms/epoch - 21ms/step
Epoch 134/500
Epoch 00134: val_loss improved from 0.01410 to 0.01371, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0243 - val_loss: 0.0137 - lr: 1.0000e-05 - 737ms/epoch - 16ms/step
Epoch 135/500
Epoch 00135: val_loss did not improve from 0.01371
45/45 - 1s - loss: 0.0259 - val_loss: 0.0138 - lr: 1.0000e-05 - 737ms/epoch - 16ms/step
Epoch 136/500
Epoch 00136: val_loss improved from 0.01371 to 0.01339, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0256 - val_loss: 0.0134 - lr: 1.0000e-05 - 772ms/epoch - 17ms/step
Epoch 137/500
Epoch 00137: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0269 - val_loss: 0.0143 - lr: 1.0000e-05 - 692ms/epoch - 15ms/step
Epoch 138/500
Epoch 00138: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0249 - val_loss: 0.0156 - lr: 1.0000e-05 - 703ms/epoch - 16ms/step
Epoch 139/500
Epoch 00139: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0262 - val_loss: 0.0149 - lr: 1.0000e-05 - 702ms/epoch - 16ms/step
Epoch 140/500
Epoch 00140: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0244 - val_loss: 0.0155 - lr: 1.0000e-05 - 724ms/epoch - 16ms/step
Epoch 141/500
Epoch 00141: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0263 - val_loss: 0.0173 - lr: 1.0000e-05 - 717ms/epoch - 16ms/step
Epoch 142/500
Epoch 00142: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0269 - val_loss: 0.0175 - lr: 1.0000e-05 - 738ms/epoch - 16ms/step
Epoch 143/500
Epoch 00143: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0258 - val_loss: 0.0169 - lr: 1.0000e-05 - 716ms/epoch - 16ms/step
Epoch 144/500
Epoch 00144: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0248 - val_loss: 0.0161 - lr: 1.0000e-05 - 736ms/epoch - 16ms/step
Epoch 145/500
Epoch 00145: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0242 - val_loss: 0.0146 - lr: 1.0000e-05 - 734ms/epoch - 16ms/step
Epoch 146/500
Epoch 00146: val_loss did not improve from 0.01339
45/45 - 1s - loss: 0.0253 - val_loss: 0.0137 - lr: 1.0000e-05 - 768ms/epoch - 17ms/step
Epoch 147/500
Epoch 00147: val_loss improved from 0.01339 to 0.01330, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0269 - val_loss: 0.0133 - lr: 1.0000e-05 - 775ms/epoch - 17ms/step
Epoch 148/500
Epoch 00148: val_loss improved from 0.01330 to 0.01313, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0252 - val_loss: 0.0131 - lr: 1.0000e-05 - 728ms/epoch - 16ms/step
Epoch 149/500
Epoch 00149: val_loss improved from 0.01313 to 0.01251, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0263 - val_loss: 0.0125 - lr: 1.0000e-05 - 788ms/epoch - 18ms/step
Epoch 150/500
Epoch 00150: val_loss did not improve from 0.01251
45/45 - 1s - loss: 0.0222 - val_loss: 0.0129 - lr: 1.0000e-05 - 724ms/epoch - 16ms/step
Epoch 151/500
Epoch 00151: val_loss improved from 0.01251 to 0.01186, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0274 - val_loss: 0.0119 - lr: 1.0000e-05 - 762ms/epoch - 17ms/step
Epoch 152/500
Epoch 00152: val_loss improved from 0.01186 to 0.01170, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0272 - val_loss: 0.0117 - lr: 1.0000e-05 - 737ms/epoch - 16ms/step
Epoch 153/500
Epoch 00153: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0253 - val_loss: 0.0123 - lr: 1.0000e-05 - 719ms/epoch - 16ms/step
Epoch 154/500
Epoch 00154: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0228 - val_loss: 0.0118 - lr: 1.0000e-05 - 727ms/epoch - 16ms/step
Epoch 155/500
Epoch 00155: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0253 - val_loss: 0.0122 - lr: 1.0000e-05 - 699ms/epoch - 16ms/step
Epoch 156/500
Epoch 00156: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0238 - val_loss: 0.0132 - lr: 1.0000e-05 - 724ms/epoch - 16ms/step
Epoch 157/500
Epoch 00157: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0234 - val_loss: 0.0141 - lr: 1.0000e-05 - 700ms/epoch - 16ms/step
Epoch 158/500
Epoch 00158: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0259 - val_loss: 0.0138 - lr: 1.0000e-05 - 776ms/epoch - 17ms/step
Epoch 159/500
Epoch 00159: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0258 - val_loss: 0.0139 - lr: 1.0000e-05 - 736ms/epoch - 16ms/step
Epoch 160/500
Epoch 00160: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0267 - val_loss: 0.0135 - lr: 1.0000e-05 - 710ms/epoch - 16ms/step
Epoch 161/500
Epoch 00161: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0226 - val_loss: 0.0142 - lr: 1.0000e-05 - 743ms/epoch - 17ms/step
Epoch 162/500
Epoch 00162: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0253 - val_loss: 0.0146 - lr: 1.0000e-05 - 750ms/epoch - 17ms/step
Epoch 163/500
Epoch 00163: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0259 - val_loss: 0.0145 - lr: 1.0000e-05 - 732ms/epoch - 16ms/step
Epoch 164/500
Epoch 00164: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0264 - val_loss: 0.0147 - lr: 1.0000e-05 - 740ms/epoch - 16ms/step
Epoch 165/500
Epoch 00165: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0237 - val_loss: 0.0141 - lr: 1.0000e-05 - 748ms/epoch - 17ms/step
Epoch 166/500
Epoch 00166: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0266 - val_loss: 0.0141 - lr: 1.0000e-05 - 774ms/epoch - 17ms/step
Epoch 167/500
Epoch 00167: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0264 - val_loss: 0.0140 - lr: 1.0000e-05 - 715ms/epoch - 16ms/step
Epoch 168/500
Epoch 00168: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0266 - val_loss: 0.0136 - lr: 1.0000e-05 - 722ms/epoch - 16ms/step
Epoch 169/500
Epoch 00169: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0239 - val_loss: 0.0127 - lr: 1.0000e-05 - 729ms/epoch - 16ms/step
Epoch 170/500
Epoch 00170: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0243 - val_loss: 0.0126 - lr: 1.0000e-05 - 701ms/epoch - 16ms/step
Epoch 171/500
Epoch 00171: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0266 - val_loss: 0.0124 - lr: 1.0000e-05 - 701ms/epoch - 16ms/step
Epoch 172/500
Epoch 00172: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0253 - val_loss: 0.0128 - lr: 1.0000e-05 - 696ms/epoch - 15ms/step
Epoch 173/500
Epoch 00173: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0242 - val_loss: 0.0133 - lr: 1.0000e-05 - 746ms/epoch - 17ms/step
Epoch 174/500
Epoch 00174: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0226 - val_loss: 0.0130 - lr: 1.0000e-05 - 730ms/epoch - 16ms/step
Epoch 175/500
Epoch 00175: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0235 - val_loss: 0.0139 - lr: 1.0000e-05 - 749ms/epoch - 17ms/step
Epoch 176/500
Epoch 00176: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0241 - val_loss: 0.0131 - lr: 1.0000e-05 - 791ms/epoch - 18ms/step
Epoch 177/500
Epoch 00177: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0228 - val_loss: 0.0127 - lr: 1.0000e-05 - 711ms/epoch - 16ms/step
Epoch 178/500
Epoch 00178: val_loss did not improve from 0.01170
45/45 - 1s - loss: 0.0246 - val_loss: 0.0128 - lr: 1.0000e-05 - 736ms/epoch - 16ms/step
Epoch 179/500
Epoch 00179: val_loss improved from 0.01170 to 0.01169, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0246 - val_loss: 0.0117 - lr: 1.0000e-05 - 959ms/epoch - 21ms/step
Epoch 180/500
Epoch 00180: val_loss improved from 0.01169 to 0.01131, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0272 - val_loss: 0.0113 - lr: 1.0000e-05 - 792ms/epoch - 18ms/step
Epoch 181/500
Epoch 00181: val_loss did not improve from 0.01131
45/45 - 1s - loss: 0.0259 - val_loss: 0.0115 - lr: 1.0000e-05 - 746ms/epoch - 17ms/step
Epoch 182/500
Epoch 00182: val_loss did not improve from 0.01131
45/45 - 1s - loss: 0.0268 - val_loss: 0.0117 - lr: 1.0000e-05 - 725ms/epoch - 16ms/step
Epoch 183/500
Epoch 00183: val_loss did not improve from 0.01131
45/45 - 1s - loss: 0.0252 - val_loss: 0.0118 - lr: 1.0000e-05 - 684ms/epoch - 15ms/step
Epoch 184/500
Epoch 00184: val_loss did not improve from 0.01131
45/45 - 1s - loss: 0.0265 - val_loss: 0.0117 - lr: 1.0000e-05 - 680ms/epoch - 15ms/step
Epoch 185/500
Epoch 00185: val_loss did not improve from 0.01131
45/45 - 1s - loss: 0.0229 - val_loss: 0.0114 - lr: 1.0000e-05 - 710ms/epoch - 16ms/step
Epoch 186/500
Epoch 00186: val_loss improved from 0.01131 to 0.01106, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0238 - val_loss: 0.0111 - lr: 1.0000e-05 - 754ms/epoch - 17ms/step
Epoch 187/500
Epoch 00187: val_loss did not improve from 0.01106
45/45 - 1s - loss: 0.0266 - val_loss: 0.0113 - lr: 1.0000e-05 - 744ms/epoch - 17ms/step
Epoch 188/500
Epoch 00188: val_loss improved from 0.01106 to 0.01062, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0267 - val_loss: 0.0106 - lr: 1.0000e-05 - 799ms/epoch - 18ms/step
Epoch 189/500
Epoch 00189: val_loss improved from 0.01062 to 0.01022, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0240 - val_loss: 0.0102 - lr: 1.0000e-05 - 753ms/epoch - 17ms/step
Epoch 190/500
Epoch 00190: val_loss improved from 0.01022 to 0.01010, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0240 - val_loss: 0.0101 - lr: 1.0000e-05 - 789ms/epoch - 18ms/step
Epoch 191/500
Epoch 00191: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0245 - val_loss: 0.0104 - lr: 1.0000e-05 - 713ms/epoch - 16ms/step
Epoch 192/500
Epoch 00192: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0250 - val_loss: 0.0109 - lr: 1.0000e-05 - 713ms/epoch - 16ms/step
Epoch 193/500
Epoch 00193: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0250 - val_loss: 0.0116 - lr: 1.0000e-05 - 733ms/epoch - 16ms/step
Epoch 194/500
Epoch 00194: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0252 - val_loss: 0.0121 - lr: 1.0000e-05 - 704ms/epoch - 16ms/step
Epoch 195/500
Epoch 00195: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0241 - val_loss: 0.0125 - lr: 1.0000e-05 - 706ms/epoch - 16ms/step
Epoch 196/500
Epoch 00196: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0275 - val_loss: 0.0138 - lr: 1.0000e-05 - 748ms/epoch - 17ms/step
Epoch 197/500
Epoch 00197: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0253 - val_loss: 0.0125 - lr: 1.0000e-05 - 736ms/epoch - 16ms/step
Epoch 198/500
Epoch 00198: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0268 - val_loss: 0.0114 - lr: 1.0000e-05 - 716ms/epoch - 16ms/step
Epoch 199/500
Epoch 00199: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0241 - val_loss: 0.0109 - lr: 1.0000e-05 - 704ms/epoch - 16ms/step
Epoch 200/500
Epoch 00200: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0238 - val_loss: 0.0109 - lr: 1.0000e-05 - 719ms/epoch - 16ms/step
Epoch 201/500
Epoch 00201: val_loss did not improve from 0.01010
45/45 - 1s - loss: 0.0282 - val_loss: 0.0105 - lr: 1.0000e-05 - 695ms/epoch - 15ms/step
Epoch 202/500
Epoch 00202: val_loss improved from 0.01010 to 0.01004, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0242 - val_loss: 0.0100 - lr: 1.0000e-05 - 894ms/epoch - 20ms/step
Epoch 203/500
Epoch 00203: val_loss did not improve from 0.01004
45/45 - 1s - loss: 0.0227 - val_loss: 0.0106 - lr: 1.0000e-05 - 703ms/epoch - 16ms/step
Epoch 204/500
Epoch 00204: val_loss did not improve from 0.01004
45/45 - 1s - loss: 0.0241 - val_loss: 0.0110 - lr: 1.0000e-05 - 733ms/epoch - 16ms/step
Epoch 205/500
Epoch 00205: val_loss did not improve from 0.01004
45/45 - 1s - loss: 0.0239 - val_loss: 0.0111 - lr: 1.0000e-05 - 729ms/epoch - 16ms/step
Epoch 206/500
Epoch 00206: val_loss did not improve from 0.01004
45/45 - 1s - loss: 0.0249 - val_loss: 0.0108 - lr: 1.0000e-05 - 720ms/epoch - 16ms/step
Epoch 207/500
Epoch 00207: val_loss did not improve from 0.01004
45/45 - 1s - loss: 0.0230 - val_loss: 0.0105 - lr: 1.0000e-05 - 747ms/epoch - 17ms/step
Epoch 208/500
Epoch 00208: val_loss improved from 0.01004 to 0.01003, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0240 - val_loss: 0.0100 - lr: 1.0000e-05 - 753ms/epoch - 17ms/step
Epoch 209/500
Epoch 00209: val_loss improved from 0.01003 to 0.00960, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0271 - val_loss: 0.0096 - lr: 1.0000e-05 - 772ms/epoch - 17ms/step
Epoch 210/500
Epoch 00210: val_loss improved from 0.00960 to 0.00936, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0235 - val_loss: 0.0094 - lr: 1.0000e-05 - 731ms/epoch - 16ms/step
Epoch 211/500
Epoch 00211: val_loss improved from 0.00936 to 0.00893, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0237 - val_loss: 0.0089 - lr: 1.0000e-05 - 752ms/epoch - 17ms/step
Epoch 212/500
Epoch 00212: val_loss improved from 0.00893 to 0.00879, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0209 - val_loss: 0.0088 - lr: 1.0000e-05 - 781ms/epoch - 17ms/step
Epoch 213/500
Epoch 00213: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0227 - val_loss: 0.0091 - lr: 1.0000e-05 - 730ms/epoch - 16ms/step
Epoch 214/500
Epoch 00214: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0222 - val_loss: 0.0092 - lr: 1.0000e-05 - 754ms/epoch - 17ms/step
Epoch 215/500
Epoch 00215: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0230 - val_loss: 0.0094 - lr: 1.0000e-05 - 702ms/epoch - 16ms/step
Epoch 216/500
Epoch 00216: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0245 - val_loss: 0.0090 - lr: 1.0000e-05 - 725ms/epoch - 16ms/step
Epoch 217/500
Epoch 00217: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0241 - val_loss: 0.0090 - lr: 1.0000e-05 - 725ms/epoch - 16ms/step
Epoch 218/500
Epoch 00218: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0237 - val_loss: 0.0092 - lr: 1.0000e-05 - 686ms/epoch - 15ms/step
Epoch 219/500
Epoch 00219: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0230 - val_loss: 0.0089 - lr: 1.0000e-05 - 736ms/epoch - 16ms/step
Epoch 220/500
Epoch 00220: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0247 - val_loss: 0.0094 - lr: 1.0000e-05 - 709ms/epoch - 16ms/step
Epoch 221/500
Epoch 00221: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0225 - val_loss: 0.0091 - lr: 1.0000e-05 - 742ms/epoch - 16ms/step
Epoch 222/500
Epoch 00222: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0234 - val_loss: 0.0090 - lr: 1.0000e-05 - 760ms/epoch - 17ms/step
Epoch 223/500
Epoch 00223: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0242 - val_loss: 0.0091 - lr: 1.0000e-05 - 783ms/epoch - 17ms/step
Epoch 224/500
Epoch 00224: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0204 - val_loss: 0.0099 - lr: 1.0000e-05 - 744ms/epoch - 17ms/step
Epoch 225/500
Epoch 00225: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0229 - val_loss: 0.0099 - lr: 1.0000e-05 - 722ms/epoch - 16ms/step
Epoch 226/500
Epoch 00226: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0240 - val_loss: 0.0093 - lr: 1.0000e-05 - 737ms/epoch - 16ms/step
Epoch 227/500
Epoch 00227: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0229 - val_loss: 0.0093 - lr: 1.0000e-05 - 742ms/epoch - 16ms/step
Epoch 228/500
Epoch 00228: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0217 - val_loss: 0.0095 - lr: 1.0000e-05 - 683ms/epoch - 15ms/step
Epoch 229/500
Epoch 00229: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0214 - val_loss: 0.0097 - lr: 1.0000e-05 - 735ms/epoch - 16ms/step
Epoch 230/500
Epoch 00230: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0242 - val_loss: 0.0097 - lr: 1.0000e-05 - 753ms/epoch - 17ms/step
Epoch 231/500
Epoch 00231: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0215 - val_loss: 0.0100 - lr: 1.0000e-05 - 726ms/epoch - 16ms/step
Epoch 232/500
Epoch 00232: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0257 - val_loss: 0.0097 - lr: 1.0000e-05 - 714ms/epoch - 16ms/step
Epoch 233/500
Epoch 00233: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0219 - val_loss: 0.0096 - lr: 1.0000e-05 - 715ms/epoch - 16ms/step
Epoch 234/500
Epoch 00234: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0238 - val_loss: 0.0098 - lr: 1.0000e-05 - 737ms/epoch - 16ms/step
Epoch 235/500
Epoch 00235: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0254 - val_loss: 0.0101 - lr: 1.0000e-05 - 724ms/epoch - 16ms/step
Epoch 236/500
Epoch 00236: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0219 - val_loss: 0.0101 - lr: 1.0000e-05 - 737ms/epoch - 16ms/step
Epoch 237/500
Epoch 00237: val_loss did not improve from 0.00879
45/45 - 1s - loss: 0.0236 - val_loss: 0.0095 - lr: 1.0000e-05 - 756ms/epoch - 17ms/step
Epoch 238/500
Epoch 00238: val_loss improved from 0.00879 to 0.00874, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0232 - val_loss: 0.0087 - lr: 1.0000e-05 - 913ms/epoch - 20ms/step
Epoch 239/500
Epoch 00239: val_loss did not improve from 0.00874
45/45 - 1s - loss: 0.0206 - val_loss: 0.0088 - lr: 1.0000e-05 - 764ms/epoch - 17ms/step
Epoch 240/500
Epoch 00240: val_loss did not improve from 0.00874
45/45 - 1s - loss: 0.0215 - val_loss: 0.0090 - lr: 1.0000e-05 - 730ms/epoch - 16ms/step
Epoch 241/500
Epoch 00241: val_loss did not improve from 0.00874
45/45 - 1s - loss: 0.0231 - val_loss: 0.0089 - lr: 1.0000e-05 - 711ms/epoch - 16ms/step
Epoch 242/500
Epoch 00242: val_loss improved from 0.00874 to 0.00852, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0252 - val_loss: 0.0085 - lr: 1.0000e-05 - 720ms/epoch - 16ms/step
Epoch 243/500
Epoch 00243: val_loss did not improve from 0.00852
45/45 - 1s - loss: 0.0241 - val_loss: 0.0086 - lr: 1.0000e-05 - 719ms/epoch - 16ms/step
Epoch 244/500
Epoch 00244: val_loss improved from 0.00852 to 0.00849, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0232 - val_loss: 0.0085 - lr: 1.0000e-05 - 777ms/epoch - 17ms/step
Epoch 245/500
Epoch 00245: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0238 - val_loss: 0.0093 - lr: 1.0000e-05 - 754ms/epoch - 17ms/step
Epoch 246/500
Epoch 00246: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0225 - val_loss: 0.0095 - lr: 1.0000e-05 - 707ms/epoch - 16ms/step
Epoch 247/500
Epoch 00247: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0214 - val_loss: 0.0097 - lr: 1.0000e-05 - 742ms/epoch - 16ms/step
Epoch 248/500
Epoch 00248: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0213 - val_loss: 0.0094 - lr: 1.0000e-05 - 752ms/epoch - 17ms/step
Epoch 249/500
Epoch 00249: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0250 - val_loss: 0.0092 - lr: 1.0000e-05 - 715ms/epoch - 16ms/step
Epoch 250/500
Epoch 00250: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0204 - val_loss: 0.0086 - lr: 1.0000e-05 - 739ms/epoch - 16ms/step
Epoch 251/500
Epoch 00251: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0222 - val_loss: 0.0086 - lr: 1.0000e-05 - 719ms/epoch - 16ms/step
Epoch 252/500
Epoch 00252: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0226 - val_loss: 0.0090 - lr: 1.0000e-05 - 741ms/epoch - 16ms/step
Epoch 253/500
Epoch 00253: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0213 - val_loss: 0.0099 - lr: 1.0000e-05 - 735ms/epoch - 16ms/step
Epoch 254/500
Epoch 00254: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0224 - val_loss: 0.0106 - lr: 1.0000e-05 - 721ms/epoch - 16ms/step
Epoch 255/500
Epoch 00255: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0239 - val_loss: 0.0097 - lr: 1.0000e-05 - 709ms/epoch - 16ms/step
Epoch 256/500
Epoch 00256: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0220 - val_loss: 0.0096 - lr: 1.0000e-05 - 723ms/epoch - 16ms/step
Epoch 257/500
Epoch 00257: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0239 - val_loss: 0.0096 - lr: 1.0000e-05 - 725ms/epoch - 16ms/step
Epoch 258/500
Epoch 00258: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0222 - val_loss: 0.0096 - lr: 1.0000e-05 - 784ms/epoch - 17ms/step
Epoch 259/500
Epoch 00259: val_loss did not improve from 0.00849
45/45 - 1s - loss: 0.0213 - val_loss: 0.0090 - lr: 1.0000e-05 - 763ms/epoch - 17ms/step
Epoch 260/500
Epoch 00260: val_loss improved from 0.00849 to 0.00825, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0219 - val_loss: 0.0082 - lr: 1.0000e-05 - 891ms/epoch - 20ms/step
Epoch 261/500
Epoch 00261: val_loss improved from 0.00825 to 0.00814, saving model to LSTM1.h5
45/45 - 1s - loss: 0.0210 - val_loss: 0.0081 - lr: 1.0000e-05 - 716ms/epoch - 16ms/step
Epoch 262/500
Epoch 00262: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0214 - val_loss: 0.0082 - lr: 1.0000e-05 - 745ms/epoch - 17ms/step
Epoch 263/500
Epoch 00263: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0224 - val_loss: 0.0082 - lr: 1.0000e-05 - 694ms/epoch - 15ms/step
Epoch 264/500
Epoch 00264: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0201 - val_loss: 0.0086 - lr: 1.0000e-05 - 741ms/epoch - 16ms/step
Epoch 265/500
Epoch 00265: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0216 - val_loss: 0.0087 - lr: 1.0000e-05 - 710ms/epoch - 16ms/step
Epoch 266/500
Epoch 00266: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0217 - val_loss: 0.0083 - lr: 1.0000e-05 - 747ms/epoch - 17ms/step
Epoch 267/500
Epoch 00267: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0204 - val_loss: 0.0082 - lr: 1.0000e-05 - 774ms/epoch - 17ms/step
Epoch 268/500
Epoch 00268: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0229 - val_loss: 0.0085 - lr: 1.0000e-05 - 692ms/epoch - 15ms/step
Epoch 269/500
Epoch 00269: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0219 - val_loss: 0.0087 - lr: 1.0000e-05 - 788ms/epoch - 18ms/step
Epoch 270/500
Epoch 00270: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0200 - val_loss: 0.0087 - lr: 1.0000e-05 - 722ms/epoch - 16ms/step
Epoch 271/500
Epoch 00271: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0212 - val_loss: 0.0083 - lr: 1.0000e-05 - 737ms/epoch - 16ms/step
Epoch 272/500
Epoch 00272: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0233 - val_loss: 0.0082 - lr: 1.0000e-05 - 723ms/epoch - 16ms/step
Epoch 273/500
Epoch 00273: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0203 - val_loss: 0.0082 - lr: 1.0000e-05 - 733ms/epoch - 16ms/step
Epoch 274/500
Epoch 00274: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0199 - val_loss: 0.0085 - lr: 1.0000e-05 - 725ms/epoch - 16ms/step
Epoch 275/500
Epoch 00275: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0205 - val_loss: 0.0085 - lr: 1.0000e-05 - 753ms/epoch - 17ms/step
Epoch 276/500
Epoch 00276: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0227 - val_loss: 0.0086 - lr: 1.0000e-05 - 690ms/epoch - 15ms/step
Epoch 277/500
Epoch 00277: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0213 - val_loss: 0.0091 - lr: 1.0000e-05 - 725ms/epoch - 16ms/step
Epoch 278/500
Epoch 00278: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0204 - val_loss: 0.0089 - lr: 1.0000e-05 - 748ms/epoch - 17ms/step
Epoch 279/500
Epoch 00279: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0201 - val_loss: 0.0090 - lr: 1.0000e-05 - 726ms/epoch - 16ms/step
Epoch 280/500
Epoch 00280: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0219 - val_loss: 0.0089 - lr: 1.0000e-05 - 792ms/epoch - 18ms/step
Epoch 281/500
Epoch 00281: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0214 - val_loss: 0.0087 - lr: 1.0000e-05 - 754ms/epoch - 17ms/step
Epoch 282/500
Epoch 00282: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0210 - val_loss: 0.0085 - lr: 1.0000e-05 - 710ms/epoch - 16ms/step
Epoch 283/500
Epoch 00283: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0232 - val_loss: 0.0087 - lr: 1.0000e-05 - 771ms/epoch - 17ms/step
Epoch 284/500
Epoch 00284: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0227 - val_loss: 0.0085 - lr: 1.0000e-05 - 743ms/epoch - 17ms/step
Epoch 285/500
Epoch 00285: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0216 - val_loss: 0.0085 - lr: 1.0000e-05 - 727ms/epoch - 16ms/step
Epoch 286/500
Epoch 00286: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0199 - val_loss: 0.0084 - lr: 1.0000e-05 - 712ms/epoch - 16ms/step
Epoch 287/500
Epoch 00287: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0204 - val_loss: 0.0086 - lr: 1.0000e-05 - 754ms/epoch - 17ms/step
Epoch 288/500
Epoch 00288: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0221 - val_loss: 0.0085 - lr: 1.0000e-05 - 697ms/epoch - 15ms/step
Epoch 289/500
Epoch 00289: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0219 - val_loss: 0.0086 - lr: 1.0000e-05 - 749ms/epoch - 17ms/step
Epoch 290/500
Epoch 00290: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0221 - val_loss: 0.0085 - lr: 1.0000e-05 - 735ms/epoch - 16ms/step
Epoch 291/500
Epoch 00291: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0204 - val_loss: 0.0086 - lr: 1.0000e-05 - 694ms/epoch - 15ms/step
Epoch 292/500
Epoch 00292: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0229 - val_loss: 0.0085 - lr: 1.0000e-05 - 732ms/epoch - 16ms/step
Epoch 293/500
Epoch 00293: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0219 - val_loss: 0.0084 - lr: 1.0000e-05 - 734ms/epoch - 16ms/step
Epoch 294/500
Epoch 00294: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0212 - val_loss: 0.0084 - lr: 1.0000e-05 - 725ms/epoch - 16ms/step
Epoch 295/500
Epoch 00295: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0215 - val_loss: 0.0088 - lr: 1.0000e-05 - 700ms/epoch - 16ms/step
Epoch 296/500
Epoch 00296: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0249 - val_loss: 0.0088 - lr: 1.0000e-05 - 717ms/epoch - 16ms/step
Epoch 297/500
Epoch 00297: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0209 - val_loss: 0.0086 - lr: 1.0000e-05 - 694ms/epoch - 15ms/step
Epoch 298/500
Epoch 00298: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0211 - val_loss: 0.0085 - lr: 1.0000e-05 - 677ms/epoch - 15ms/step
Epoch 299/500
Epoch 00299: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0239 - val_loss: 0.0085 - lr: 1.0000e-05 - 701ms/epoch - 16ms/step
Epoch 300/500
Epoch 00300: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0211 - val_loss: 0.0085 - lr: 1.0000e-05 - 707ms/epoch - 16ms/step
Epoch 301/500
Epoch 00301: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0203 - val_loss: 0.0086 - lr: 1.0000e-05 - 740ms/epoch - 16ms/step
Epoch 302/500
Epoch 00302: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0210 - val_loss: 0.0086 - lr: 1.0000e-05 - 736ms/epoch - 16ms/step
Epoch 303/500
Epoch 00303: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0218 - val_loss: 0.0087 - lr: 1.0000e-05 - 728ms/epoch - 16ms/step
Epoch 304/500
Epoch 00304: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0213 - val_loss: 0.0089 - lr: 1.0000e-05 - 743ms/epoch - 17ms/step
Epoch 305/500
Epoch 00305: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0185 - val_loss: 0.0089 - lr: 1.0000e-05 - 711ms/epoch - 16ms/step
Epoch 306/500
Epoch 00306: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0213 - val_loss: 0.0089 - lr: 1.0000e-05 - 763ms/epoch - 17ms/step
Epoch 307/500
Epoch 00307: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0195 - val_loss: 0.0090 - lr: 1.0000e-05 - 709ms/epoch - 16ms/step
Epoch 308/500
Epoch 00308: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0231 - val_loss: 0.0090 - lr: 1.0000e-05 - 721ms/epoch - 16ms/step
Epoch 309/500
Epoch 00309: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0210 - val_loss: 0.0089 - lr: 1.0000e-05 - 769ms/epoch - 17ms/step
Epoch 310/500
Epoch 00310: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0216 - val_loss: 0.0090 - lr: 1.0000e-05 - 749ms/epoch - 17ms/step
Epoch 311/500
Epoch 00311: val_loss did not improve from 0.00814
45/45 - 1s - loss: 0.0218 - val_loss: 0.0090 - lr: 1.0000e-05 - 701ms/epoch - 16ms/step
Epoch 00311: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 216.26215770312788
RMSE: 14.705854538350632
MAPE: 11.92367463073216
EMA
Prediction vs Close: 54.1% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 122.96033735804009
RMSE: 11.08874823224155
MAPE: 9.251696034357076
WMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 56.80196950717294
RMSE: 7.5367081346681415
MAPE: 5.956122993340066
DEMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 42.16% Accuracy
MSE: 120.51656357671082
RMSE: 10.978003624371365
MAPE: 9.343426819843298
KAMA
Prediction vs Close: 56.34% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 80.73205723036311
RMSE: 8.985101959931402
MAPE: 7.079003376879244
MIDPOINT
MIDPOINT([input_arrays], [timeperiod=14])
MidPoint over period (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 14
Outputs:
real
14
Working on MIDPOINT predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.53 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4212.289, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3747.746, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.34 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3523.401, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3387.759, Time=0.13 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.62 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.17 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3389.758, Time=0.28 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.274 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1689.879
Date: Sun, 12 Dec 2021 AIC 3387.759
Time: 12:09:13 BIC 3406.522
Sample: 0 HQIC 3394.964
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1878 0.003 -345.315 0.000 -1.195 -1.181
ar.L2 -0.8876 0.007 -121.809 0.000 -0.902 -0.873
ar.L3 -0.3957 0.007 -60.127 0.000 -0.409 -0.383
sigma2 3.8904 0.020 193.404 0.000 3.851 3.930
===================================================================================
Ljung-Box (L1) (Q): 13.21 Jarque-Bera (JB): 1659080.01
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.08 Skew: 3.28
Prob(H) (two-sided): 0.00 Kurtosis: 225.31
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
WARNING:tensorflow:Layer lstm_5 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.06587, saving model to LSTM1.h5
58/58 - 3s - loss: 0.1860 - val_loss: 0.0659 - lr: 0.0010 - 3s/epoch - 48ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.06587
58/58 - 1s - loss: 0.0798 - val_loss: 0.0807 - lr: 0.0010 - 910ms/epoch - 16ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.06587
58/58 - 1s - loss: 0.0675 - val_loss: 0.4197 - lr: 0.0010 - 876ms/epoch - 15ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.06587 to 0.01530, saving model to LSTM1.h5
58/58 - 1s - loss: 0.0426 - val_loss: 0.0153 - lr: 0.0010 - 939ms/epoch - 16ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.01530
58/58 - 1s - loss: 0.0399 - val_loss: 0.2015 - lr: 0.0010 - 959ms/epoch - 17ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.01530 to 0.00621, saving model to LSTM1.h5
58/58 - 1s - loss: 0.0366 - val_loss: 0.0062 - lr: 0.0010 - 933ms/epoch - 16ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0393 - val_loss: 0.0567 - lr: 0.0010 - 885ms/epoch - 15ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0373 - val_loss: 0.0068 - lr: 0.0010 - 911ms/epoch - 16ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0338 - val_loss: 0.2309 - lr: 0.0010 - 870ms/epoch - 15ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0301 - val_loss: 0.0215 - lr: 0.0010 - 926ms/epoch - 16ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00011: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0308 - val_loss: 0.0065 - lr: 0.0010 - 890ms/epoch - 15ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0351 - val_loss: 0.0083 - lr: 1.0000e-04 - 866ms/epoch - 15ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0251 - val_loss: 0.0091 - lr: 1.0000e-04 - 854ms/epoch - 15ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0267 - val_loss: 0.0091 - lr: 1.0000e-04 - 899ms/epoch - 15ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0248 - val_loss: 0.0081 - lr: 1.0000e-04 - 867ms/epoch - 15ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00016: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0263 - val_loss: 0.0078 - lr: 1.0000e-04 - 899ms/epoch - 15ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0220 - val_loss: 0.0080 - lr: 1.0000e-05 - 923ms/epoch - 16ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0214 - val_loss: 0.0079 - lr: 1.0000e-05 - 842ms/epoch - 15ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0247 - val_loss: 0.0078 - lr: 1.0000e-05 - 850ms/epoch - 15ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0233 - val_loss: 0.0077 - lr: 1.0000e-05 - 853ms/epoch - 15ms/step
Epoch 21/500
Epoch 00021: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00021: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0254 - val_loss: 0.0079 - lr: 1.0000e-05 - 882ms/epoch - 15ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0238 - val_loss: 0.0080 - lr: 1.0000e-05 - 882ms/epoch - 15ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0243 - val_loss: 0.0079 - lr: 1.0000e-05 - 853ms/epoch - 15ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0240 - val_loss: 0.0079 - lr: 1.0000e-05 - 892ms/epoch - 15ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0228 - val_loss: 0.0077 - lr: 1.0000e-05 - 880ms/epoch - 15ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0242 - val_loss: 0.0079 - lr: 1.0000e-05 - 844ms/epoch - 15ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0238 - val_loss: 0.0081 - lr: 1.0000e-05 - 859ms/epoch - 15ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0233 - val_loss: 0.0080 - lr: 1.0000e-05 - 883ms/epoch - 15ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0213 - val_loss: 0.0080 - lr: 1.0000e-05 - 898ms/epoch - 15ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0209 - val_loss: 0.0080 - lr: 1.0000e-05 - 887ms/epoch - 15ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0219 - val_loss: 0.0081 - lr: 1.0000e-05 - 875ms/epoch - 15ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0245 - val_loss: 0.0081 - lr: 1.0000e-05 - 877ms/epoch - 15ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0219 - val_loss: 0.0080 - lr: 1.0000e-05 - 836ms/epoch - 14ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0226 - val_loss: 0.0079 - lr: 1.0000e-05 - 869ms/epoch - 15ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0225 - val_loss: 0.0079 - lr: 1.0000e-05 - 899ms/epoch - 15ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0265 - val_loss: 0.0080 - lr: 1.0000e-05 - 887ms/epoch - 15ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0210 - val_loss: 0.0080 - lr: 1.0000e-05 - 857ms/epoch - 15ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0246 - val_loss: 0.0082 - lr: 1.0000e-05 - 863ms/epoch - 15ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0216 - val_loss: 0.0083 - lr: 1.0000e-05 - 878ms/epoch - 15ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0222 - val_loss: 0.0079 - lr: 1.0000e-05 - 844ms/epoch - 15ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0237 - val_loss: 0.0078 - lr: 1.0000e-05 - 857ms/epoch - 15ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0237 - val_loss: 0.0078 - lr: 1.0000e-05 - 868ms/epoch - 15ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0247 - val_loss: 0.0076 - lr: 1.0000e-05 - 876ms/epoch - 15ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0244 - val_loss: 0.0075 - lr: 1.0000e-05 - 897ms/epoch - 15ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0243 - val_loss: 0.0077 - lr: 1.0000e-05 - 865ms/epoch - 15ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0229 - val_loss: 0.0078 - lr: 1.0000e-05 - 916ms/epoch - 16ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0207 - val_loss: 0.0076 - lr: 1.0000e-05 - 874ms/epoch - 15ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0248 - val_loss: 0.0074 - lr: 1.0000e-05 - 878ms/epoch - 15ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0226 - val_loss: 0.0073 - lr: 1.0000e-05 - 881ms/epoch - 15ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0219 - val_loss: 0.0075 - lr: 1.0000e-05 - 897ms/epoch - 15ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0221 - val_loss: 0.0071 - lr: 1.0000e-05 - 852ms/epoch - 15ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0223 - val_loss: 0.0074 - lr: 1.0000e-05 - 866ms/epoch - 15ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0221 - val_loss: 0.0077 - lr: 1.0000e-05 - 841ms/epoch - 15ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0241 - val_loss: 0.0075 - lr: 1.0000e-05 - 847ms/epoch - 15ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0235 - val_loss: 0.0075 - lr: 1.0000e-05 - 851ms/epoch - 15ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00621
58/58 - 1s - loss: 0.0200 - val_loss: 0.0075 - lr: 1.0000e-05 - 870ms/epoch - 15ms/step
Epoch 00056: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 216.26215770312788
RMSE: 14.705854538350632
MAPE: 11.92367463073216
EMA
Prediction vs Close: 54.1% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 122.96033735804009
RMSE: 11.08874823224155
MAPE: 9.251696034357076
WMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 56.80196950717294
RMSE: 7.5367081346681415
MAPE: 5.956122993340066
DEMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 42.16% Accuracy
MSE: 120.51656357671082
RMSE: 10.978003624371365
MAPE: 9.343426819843298
KAMA
Prediction vs Close: 56.34% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 80.73205723036311
RMSE: 8.985101959931402
MAPE: 7.079003376879244
MIDPOINT
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 45.9% Accuracy
MSE: 75.80130571921515
RMSE: 8.70639453041356
MAPE: 7.130945881105426
T3
T3([input_arrays], [timeperiod=5], [vfactor=0.7])
Triple Exponential Moving Average (T3) (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 5
vfactor: 0.7
Outputs:
real
19
Working on T3 predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.55 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4414.515, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3944.062, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.52 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3715.173, Time=0.08 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3577.471, Time=0.11 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.88 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.80 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3579.471, Time=0.25 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.307 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1784.736
Date: Sun, 12 Dec 2021 AIC 3577.471
Time: 12:11:36 BIC 3596.235
Sample: 0 HQIC 3584.677
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1982 0.003 -389.844 0.000 -1.204 -1.192
ar.L2 -0.8974 0.006 -139.861 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.862 0.000 -0.410 -0.387
sigma2 4.9242 0.023 215.469 0.000 4.879 4.969
===================================================================================
Ljung-Box (L1) (Q): 14.55 Jarque-Bera (JB): 2468024.38
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 274.15
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
WARNING:tensorflow:Layer lstm_6 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.05236, saving model to LSTM1.h5
43/43 - 3s - loss: 0.1928 - val_loss: 0.0524 - lr: 0.0010 - 3s/epoch - 61ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.1446 - val_loss: 0.0525 - lr: 0.0010 - 704ms/epoch - 16ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.1028 - val_loss: 0.9645 - lr: 0.0010 - 634ms/epoch - 15ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0612 - val_loss: 0.2236 - lr: 0.0010 - 657ms/epoch - 15ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0534 - val_loss: 0.1234 - lr: 0.0010 - 688ms/epoch - 16ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0439 - val_loss: 0.0788 - lr: 0.0010 - 668ms/epoch - 16ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0362 - val_loss: 0.0777 - lr: 1.0000e-04 - 694ms/epoch - 16ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0359 - val_loss: 0.0834 - lr: 1.0000e-04 - 700ms/epoch - 16ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0339 - val_loss: 0.0859 - lr: 1.0000e-04 - 650ms/epoch - 15ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0323 - val_loss: 0.0797 - lr: 1.0000e-04 - 712ms/epoch - 17ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0311 - val_loss: 0.0836 - lr: 1.0000e-04 - 720ms/epoch - 17ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0294 - val_loss: 0.0836 - lr: 1.0000e-05 - 706ms/epoch - 16ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0328 - val_loss: 0.0842 - lr: 1.0000e-05 - 677ms/epoch - 16ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0290 - val_loss: 0.0851 - lr: 1.0000e-05 - 701ms/epoch - 16ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0294 - val_loss: 0.0855 - lr: 1.0000e-05 - 636ms/epoch - 15ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0311 - val_loss: 0.0847 - lr: 1.0000e-05 - 659ms/epoch - 15ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0303 - val_loss: 0.0844 - lr: 1.0000e-05 - 705ms/epoch - 16ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0310 - val_loss: 0.0839 - lr: 1.0000e-05 - 711ms/epoch - 17ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0346 - val_loss: 0.0832 - lr: 1.0000e-05 - 700ms/epoch - 16ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0296 - val_loss: 0.0837 - lr: 1.0000e-05 - 720ms/epoch - 17ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0314 - val_loss: 0.0832 - lr: 1.0000e-05 - 676ms/epoch - 16ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0309 - val_loss: 0.0828 - lr: 1.0000e-05 - 657ms/epoch - 15ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0310 - val_loss: 0.0832 - lr: 1.0000e-05 - 684ms/epoch - 16ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0297 - val_loss: 0.0835 - lr: 1.0000e-05 - 685ms/epoch - 16ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0330 - val_loss: 0.0843 - lr: 1.0000e-05 - 600ms/epoch - 14ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0317 - val_loss: 0.0855 - lr: 1.0000e-05 - 654ms/epoch - 15ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0303 - val_loss: 0.0864 - lr: 1.0000e-05 - 634ms/epoch - 15ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0290 - val_loss: 0.0858 - lr: 1.0000e-05 - 618ms/epoch - 14ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0299 - val_loss: 0.0857 - lr: 1.0000e-05 - 661ms/epoch - 15ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0284 - val_loss: 0.0850 - lr: 1.0000e-05 - 644ms/epoch - 15ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0326 - val_loss: 0.0849 - lr: 1.0000e-05 - 662ms/epoch - 15ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0331 - val_loss: 0.0842 - lr: 1.0000e-05 - 631ms/epoch - 15ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0312 - val_loss: 0.0845 - lr: 1.0000e-05 - 680ms/epoch - 16ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0287 - val_loss: 0.0853 - lr: 1.0000e-05 - 671ms/epoch - 16ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0303 - val_loss: 0.0844 - lr: 1.0000e-05 - 671ms/epoch - 16ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0276 - val_loss: 0.0851 - lr: 1.0000e-05 - 698ms/epoch - 16ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0283 - val_loss: 0.0854 - lr: 1.0000e-05 - 661ms/epoch - 15ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0301 - val_loss: 0.0857 - lr: 1.0000e-05 - 637ms/epoch - 15ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0284 - val_loss: 0.0851 - lr: 1.0000e-05 - 658ms/epoch - 15ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0290 - val_loss: 0.0847 - lr: 1.0000e-05 - 651ms/epoch - 15ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0301 - val_loss: 0.0841 - lr: 1.0000e-05 - 692ms/epoch - 16ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0274 - val_loss: 0.0851 - lr: 1.0000e-05 - 670ms/epoch - 16ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0300 - val_loss: 0.0844 - lr: 1.0000e-05 - 631ms/epoch - 15ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0284 - val_loss: 0.0844 - lr: 1.0000e-05 - 691ms/epoch - 16ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0291 - val_loss: 0.0832 - lr: 1.0000e-05 - 665ms/epoch - 15ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0273 - val_loss: 0.0833 - lr: 1.0000e-05 - 687ms/epoch - 16ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0290 - val_loss: 0.0836 - lr: 1.0000e-05 - 646ms/epoch - 15ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0290 - val_loss: 0.0844 - lr: 1.0000e-05 - 684ms/epoch - 16ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0317 - val_loss: 0.0849 - lr: 1.0000e-05 - 689ms/epoch - 16ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0267 - val_loss: 0.0838 - lr: 1.0000e-05 - 655ms/epoch - 15ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.05236
43/43 - 1s - loss: 0.0291 - val_loss: 0.0829 - lr: 1.0000e-05 - 639ms/epoch - 15ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 216.26215770312788
RMSE: 14.705854538350632
MAPE: 11.92367463073216
EMA
Prediction vs Close: 54.1% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 122.96033735804009
RMSE: 11.08874823224155
MAPE: 9.251696034357076
WMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 56.80196950717294
RMSE: 7.5367081346681415
MAPE: 5.956122993340066
DEMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 42.16% Accuracy
MSE: 120.51656357671082
RMSE: 10.978003624371365
MAPE: 9.343426819843298
KAMA
Prediction vs Close: 56.34% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 80.73205723036311
RMSE: 8.985101959931402
MAPE: 7.079003376879244
MIDPOINT
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 45.9% Accuracy
MSE: 75.80130571921515
RMSE: 8.70639453041356
MAPE: 7.130945881105426
T3
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 46.36148283886003
RMSE: 6.808926702415001
MAPE: 5.596706034896943
TEMA
TEMA([input_arrays], [timeperiod=30])
Triple Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
9
Working on TEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.64 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4352.703, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3889.412, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.38 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3689.930, Time=0.08 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3574.245, Time=0.12 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.56 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.04 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3576.245, Time=0.27 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.208 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1783.123
Date: Sun, 12 Dec 2021 AIC 3574.245
Time: 12:13:37 BIC 3593.008
Sample: 0 HQIC 3581.451
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1480 0.004 -302.430 0.000 -1.155 -1.141
ar.L2 -0.8300 0.008 -99.682 0.000 -0.846 -0.814
ar.L3 -0.3687 0.007 -50.527 0.000 -0.383 -0.354
sigma2 4.9055 0.028 175.970 0.000 4.851 4.960
===================================================================================
Ljung-Box (L1) (Q): 11.61 Jarque-Bera (JB): 1261976.58
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.16 Skew: 2.52
Prob(H) (two-sided): 0.00 Kurtosis: 196.90
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
WARNING:tensorflow:Layer lstm_7 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.01596, saving model to LSTM1.h5
90/90 - 3s - loss: 0.2644 - val_loss: 0.0160 - lr: 0.0010 - 3s/epoch - 36ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0703 - val_loss: 0.0610 - lr: 0.0010 - 1s/epoch - 15ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.1296 - val_loss: 0.0363 - lr: 0.0010 - 1s/epoch - 15ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0808 - val_loss: 1.3399 - lr: 0.0010 - 1s/epoch - 15ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0449 - val_loss: 0.2443 - lr: 0.0010 - 1s/epoch - 15ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0420 - val_loss: 0.0862 - lr: 0.0010 - 1s/epoch - 15ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0426 - val_loss: 0.1005 - lr: 1.0000e-04 - 1s/epoch - 15ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0357 - val_loss: 0.0937 - lr: 1.0000e-04 - 1s/epoch - 14ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0392 - val_loss: 0.0879 - lr: 1.0000e-04 - 1s/epoch - 15ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0381 - val_loss: 0.0818 - lr: 1.0000e-04 - 1s/epoch - 15ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0355 - val_loss: 0.0753 - lr: 1.0000e-04 - 1s/epoch - 14ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0306 - val_loss: 0.0757 - lr: 1.0000e-05 - 1s/epoch - 14ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0336 - val_loss: 0.0761 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0312 - val_loss: 0.0764 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0353 - val_loss: 0.0757 - lr: 1.0000e-05 - 1s/epoch - 16ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0306 - val_loss: 0.0754 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0323 - val_loss: 0.0756 - lr: 1.0000e-05 - 1s/epoch - 14ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0324 - val_loss: 0.0766 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0339 - val_loss: 0.0777 - lr: 1.0000e-05 - 1s/epoch - 16ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0324 - val_loss: 0.0783 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0295 - val_loss: 0.0782 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0327 - val_loss: 0.0788 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0320 - val_loss: 0.0787 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0281 - val_loss: 0.0780 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0332 - val_loss: 0.0776 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0322 - val_loss: 0.0770 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0315 - val_loss: 0.0759 - lr: 1.0000e-05 - 1s/epoch - 14ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0286 - val_loss: 0.0763 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0295 - val_loss: 0.0767 - lr: 1.0000e-05 - 1s/epoch - 16ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0312 - val_loss: 0.0764 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0332 - val_loss: 0.0753 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0289 - val_loss: 0.0756 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0312 - val_loss: 0.0762 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0313 - val_loss: 0.0745 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0345 - val_loss: 0.0749 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0318 - val_loss: 0.0762 - lr: 1.0000e-05 - 1s/epoch - 14ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0302 - val_loss: 0.0756 - lr: 1.0000e-05 - 1s/epoch - 16ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0289 - val_loss: 0.0750 - lr: 1.0000e-05 - 1s/epoch - 16ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0309 - val_loss: 0.0755 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0309 - val_loss: 0.0754 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0302 - val_loss: 0.0753 - lr: 1.0000e-05 - 1s/epoch - 14ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0326 - val_loss: 0.0736 - lr: 1.0000e-05 - 1s/epoch - 14ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0318 - val_loss: 0.0728 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0282 - val_loss: 0.0731 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0314 - val_loss: 0.0749 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0289 - val_loss: 0.0755 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0304 - val_loss: 0.0760 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0280 - val_loss: 0.0754 - lr: 1.0000e-05 - 1s/epoch - 16ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0281 - val_loss: 0.0756 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0303 - val_loss: 0.0775 - lr: 1.0000e-05 - 1s/epoch - 16ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01596
90/90 - 1s - loss: 0.0293 - val_loss: 0.0771 - lr: 1.0000e-05 - 1s/epoch - 15ms/step
Epoch 00051: early stopping
SMA Prediction vs Close: 52.61% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 216.26215770312788 RMSE: 14.705854538350632 MAPE: 11.92367463073216 EMA Prediction vs Close: 54.1% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 122.96033735804009 RMSE: 11.08874823224155 MAPE: 9.251696034357076 WMA Prediction vs Close: 53.73% Accuracy Prediction vs Prediction: 46.64% Accuracy MSE: 56.80196950717294 RMSE: 7.5367081346681415 MAPE: 5.956122993340066 DEMA Prediction vs Close: 52.99% Accuracy Prediction vs Prediction: 42.16% Accuracy MSE: 120.51656357671082 RMSE: 10.978003624371365 MAPE: 9.343426819843298 KAMA Prediction vs Close: 56.34% Accuracy Prediction vs Prediction: 49.63% Accuracy MSE: 80.73205723036311 RMSE: 8.985101959931402 MAPE: 7.079003376879244 MIDPOINT Prediction vs Close: 53.73% Accuracy Prediction vs Prediction: 45.9% Accuracy MSE: 75.80130571921515 RMSE: 8.70639453041356 MAPE: 7.130945881105426 T3 Prediction vs Close: 52.24% Accuracy Prediction vs Prediction: 46.64% Accuracy MSE: 46.36148283886003 RMSE: 6.808926702415001 MAPE: 5.596706034896943 TEMA Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 48.51% Accuracy MSE: 32.02006831549358 RMSE: 5.658627776722337 MAPE: 4.840005256997922 Runtime: mins: 20.992133364683333
Architecture used¶
In [ ]:
from google.colab import files
import cv2
uploaded = files.upload()
Saving Experiment1.png to Experiment1 (2).png
In [ ]:
img = cv2.imread('Experiment1.png')
plt.figure(figsize=(20,10))
plt.axis("off")
plt.title('LSTM Architecture '+imgfile,fontsize=18)
plt.imshow(img)
Out[ ]:
<matplotlib.image.AxesImage at 0x7fdf309b3090>
Excess kurtosis is a metric that compares the kurtosis of a distribution against the kurtosis of a normal distribution. The kurtosis of a normal distribution equals 3. Therefore, the excess kurtosis is found using the formula below:
Excess Kurtosis = Kurtosis – 3
Model Plots¶
In [ ]:
np.save("X_train_appl.npy", X_train)
np.save("y_train_appl.npy", y_train)
np.save("X_test_appl.npy", X_test)
np.save("y_test_appl.npy", y_test)
np.save("yc_train_appl.npy", yc_train)
np.save("yc_test_appl.npy", yc_test)
np.save('index_train_appl.npy', index_train)
np.save('index_test_appl.npy', index_test)
In [ ]:
list(simulation1.keys())
Out[ ]:
['SMA', 'EMA', 'WMA', 'DEMA', 'KAMA', 'MIDPOINT', 'T3', 'TEMA']
In [71]:
with open('simulation1_data.json') as json_file:
simulation1 = json.load(json_file)
fileimg = 'Experiment1'
In [72]:
for i in range(len(list(simulation1.keys()))):
SIM = list(simulation1.keys())[i]
plot_train(simulation1,SIM)
plot_test(simulation1,SIM)
----- Train RMSE for SMA ----- 7.922367251770581 ----- Train_MSE_LSTM for SMA ----- 62.763902871926945 ----- Train MAE LSTM for SMA ----- 6.766703454283121
----- Test RMSE for SMA----- 14.705854538350632 ----- Test_MSE_LSTM for SMA----- 216.26215770312788 ----- Test_MAE_LSTM for SMA----- 11.92367463073216
----- Train RMSE for EMA ----- 9.326837823498344 ----- Train_MSE_LSTM for EMA ----- 86.98990378583935 ----- Train MAE LSTM for EMA ----- 8.227210225627324
----- Test RMSE for EMA----- 11.08874823224155 ----- Test_MSE_LSTM for EMA----- 122.96033735804009 ----- Test_MAE_LSTM for EMA----- 9.251696034357076
----- Train RMSE for WMA ----- 9.567750607562207 ----- Train_MSE_LSTM for WMA ----- 91.54185168850697 ----- Train MAE LSTM for WMA ----- 8.483814814644358
----- Test RMSE for WMA----- 7.5367081346681415 ----- Test_MSE_LSTM for WMA----- 56.80196950717294 ----- Test_MAE_LSTM for WMA----- 5.956122993340066
----- Train RMSE for DEMA ----- 11.260564301359528 ----- Train_MSE_LSTM for DEMA ----- 126.80030838505259 ----- Train MAE LSTM for DEMA ----- 10.029241457055923
----- Test RMSE for DEMA----- 10.978003624371365 ----- Test_MSE_LSTM for DEMA----- 120.51656357671082 ----- Test_MAE_LSTM for DEMA----- 9.343426819843298
----- Train RMSE for KAMA ----- 9.728924164194906 ----- Train_MSE_LSTM for KAMA ----- 94.65196539265554 ----- Train MAE LSTM for KAMA ----- 8.726636092269294
----- Test RMSE for KAMA----- 8.985101959931402 ----- Test_MSE_LSTM for KAMA----- 80.73205723036311 ----- Test_MAE_LSTM for KAMA----- 7.079003376879244
----- Train RMSE for MIDPOINT ----- 8.489283494179926 ----- Train_MSE_LSTM for MIDPOINT ----- 72.06793424455573 ----- Train MAE LSTM for MIDPOINT ----- 7.510015655912938
----- Test RMSE for MIDPOINT----- 8.70639453041356 ----- Test_MSE_LSTM for MIDPOINT----- 75.80130571921515 ----- Test_MAE_LSTM for MIDPOINT----- 7.130945881105426
----- Train RMSE for T3 ----- 10.915128509512805 ----- Train_MSE_LSTM for T3 ----- 119.14003037917921 ----- Train MAE LSTM for T3 ----- 9.760355664290358
----- Test RMSE for T3----- 6.808926702415001 ----- Test_MSE_LSTM for T3----- 46.36148283886003 ----- Test_MAE_LSTM for T3----- 5.596706034896943
----- Train RMSE for TEMA ----- 6.723556059985799 ----- Train_MSE_LSTM for TEMA ----- 45.20620609177176 ----- Train MAE LSTM for TEMA ----- 4.40413747354074
----- Test RMSE for TEMA----- 5.658627776722337 ----- Test_MSE_LSTM for TEMA----- 32.02006831549358 ----- Test_MAE_LSTM for TEMA----- 4.840005256997922
Univariate Arima Multistep MutiVariate LSTM Hybrid Model Experiment 2¶
In [ ]:
def get_lstm(data,original_data, train_len, test_len,img_file,ma ,lstm_len=3):
# prepare train and test data
X_value = pd.DataFrame(data.iloc[:, :])
y_value = pd.DataFrame(data.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
# Get data and check shape
X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X)
y_train, y_test, = split_train_test(y)
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
det = 20
input_dim = X_train.shape[1]#3
feature_size = X_train.shape[2]#24
output_dim = y_train.shape[1]#1
# # Option 1
# # Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(256, activation='relu', kernel_initializer='he_normal', input_shape=(input_dim, feature_size)))
# model.add(Dense(units=64,activation='relu'))
# model.add(Dropout(0.5))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(learning_rate = 0.001), loss='mse')
# ## Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# option 2
model = Sequential()
model.add(Bidirectional(LSTM(units= 128), input_shape=(input_dim, feature_size)))
model.add(Dense(64))
model.add(Dense(units=output_dim))
model.compile(optimizer=Adam(learning_rate = 0.001), loss='mean_squared_error', metrics=['accuracy'])
# Common code
callbacks = [
EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
ModelCheckpoint('LSTM2.h5', verbose=1, save_best_only=True, save_weights_only=True)]
fname1 = img_file+'.png'
tensorflow.keras.utils.plot_model(
model, to_file=fname1, show_shapes=True, show_dtype=False,
show_layer_names=True, expand_nested=False, dpi=96,
layer_range=None, show_layer_activations=False
)
history = model.fit(X_train, y_train, epochs=500, batch_size=10, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# plot loss
fname2 = img_file+'-'+ma
plt.title(img_file+'-'+ma+' Loss')
plt.xlabel("Epochs")
plt.ylabel("Loss")
pyplot.plot(history.history['loss'], label='train')
pyplot.plot(history.history['val_loss'], label='validation')
pyplot.legend()
pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# Option 3
# define custom activation
#
# class Double_Tanh(Activation):
# def __init__(self, activation, **kwargs):
# super(Double_Tanh, self).__init__(activation, **kwargs)
# self.__name__ = 'double_tanh'
# def double_tanh(x):
# return (K.tanh(x) * 2)
# get_custom_objects().update({'double_tanh':Double_Tanh(double_tanh)})
# # Model Generation
# model = Sequential()
# #check https://machinelearningmastery.com/use-weight-regularization-lstm-networks-time-series-forecasting/
# model.add(LSTM(25, input_shape=(input_dim, feature_size), dropout=0.2, kernel_regularizer=l1_l2(0.00,0.00), bias_regularizer=l1_l2(0.00,0.00)))
# model.add(Dense(1))
# model.add(Activation(double_tanh))
# model.compile(loss='mean_squared_error', optimizer='adam', metrics=['mse', 'mae'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Option 4
# Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(units=lstm_len, return_sequences=True, input_shape=(x_train.shape[1], 1)))
# model.add(LSTM(units=int(lstm_len/2)))
# model.add(Dense(1, activation='sigmoid'))
# model.compile(loss='mean_squared_error', optimizer='adam')
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Generate predictions
predictiontr = model.predict(X_train, verbose=0)
predictiontr = y_scaler.inverse_transform(predictiontr).tolist()
outputtr = []
for i in range(len(predictiontr)):
outputtr.extend(predictiontr[i])
predictiontr = outputtr
# Generate error data
## replace with yc , xtest generated by new multistep method
mse_tr = mean_squared_error(y_train, predictiontr)
rmse_tr = mse_tr ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictiontr))
mae_tr = mean_absolute_error(y_train, pd.Series(predictiontr))
# Original_tr = pd.Series(yc_train)
Original_tr = y_scaler.inverse_transform(y_train).flatten().tolist()
predictionte = model.predict(X_test, verbose=0)
predictionte = (y_scaler.inverse_transform(predictionte)-det).tolist()
outputte = []
for i in range(len(predictionte)):
outputte.extend(predictionte[i])
predictionte = outputte
# Generate error data
mse_te = mean_squared_error(y_test, predictionte)
rmse_te = mse_te ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictionte))
mae_te = mean_absolute_error(y_test, pd.Series(predictionte))
# Original_te = pd.Series(yc_test)
Original_te = y_scaler.inverse_transform(y_test).flatten().tolist()
return Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,Original_te,predictionte, mse_te,rmse_te,mae_te
In [ ]:
if __name__ == '__main__':
start_time = timeit.default_timer()
simulation2 = {}
imgfile = 'Experiment2'
for ma in optimized_period:
print(ma)
print(functions[ma])
print ( int( optimized_period[ma]))
# if ma == 'SMA':
low_vol = df.apply(lambda c: functions[ma](c, timeperiod = int( optimized_period[ma])))
low_vol = low_vol.fillna(0)
low_vol_data = df['close']
high_vol = pd.DataFrame()
df2 = df.copy()
for i in df2.columns:
if i in low_vol.columns:
high_vol[i] = df2[i].subtract(low_vol[i], fill_value=0)
high_vol_data = df['close']
## *****************************************************
# Generate ARIMA and LSTM predictions
print('\nWorking on ' + ma + ' predictions')
try:
print('parameters used : ', train_len, test_len)
low_vol_Original, low_vol_prediction, low_vol_mse, low_vol_rmse,low_vol_mae = get_arima(low_vol,low_vol_data, train_len, test_len)
except:
print('ARIMA error, skipping to next MA type')
continue
Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,high_vol_Original, high_vol_prediction, high_vol_mse, high_vol_rmse,high_vol_mae, = get_lstm(high_vol,high_vol_data, train_len, test_len,imgfile,ma)
final_prediction_tr = df['close'].head(train_len).values + pd.Series(predictiontr) # ignoring first 3 steps
mse_ftr = mean_squared_error(df['close'].head(train_len).values,final_prediction_tr.values)
rmse_ftr = mse_ftr ** 0.5
mape_ftr = mean_absolute_percentage_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
mae_ftr = mean_absolute_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
final_prediction = pd.Series(low_vol_prediction[3:]) + pd.Series(high_vol_prediction)
mse = mean_squared_error(df['close'].tail(test_len).values,final_prediction.values)
rmse = mse ** 0.5
mape = mean_absolute_percentage_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
mae = mean_absolute_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
# Generate prediction accuracy
actual = df['close'].tail(test_len).values
result_1 = []
result_2 = []
for i in range(1, len(final_prediction)):
# Compare prediction to previous close price
if final_prediction[i] > actual[i-1] and actual[i] > actual[i-1]:
result_1.append(1)
elif final_prediction[i] < actual[i-1] and actual[i] < actual[i-1]:
result_1.append(1)
else:
result_1.append(0)
# Compare prediction to previous prediction
if final_prediction[i] > final_prediction[i-1] and actual[i] > actual[i-1]:
result_2.append(1)
elif final_prediction[i] < final_prediction[i-1] and actual[i] < actual[i-1]:
result_2.append(1)
else:
result_2.append(0)
accuracy_1 = np.mean(result_1)
accuracy_2 = np.mean(result_2)
simulation2[ma] = {'low_vol': {'original':list(low_vol_Original), 'prediction': list(low_vol_prediction) , 'mse': low_vol_mse,
'rmse': low_vol_rmse, 'mae' : low_vol_mae},
'high_vol': {'original':list(high_vol_Original),'prediction': list(high_vol_prediction), 'mse': high_vol_mse,
'rmse': high_vol_rmse, 'mae' : high_vol_mae},
'final_tr': {'original':df['close'].head(train_len).tolist(),'prediction': final_prediction_tr.values.tolist(), 'mse': mse_ftr,
'rmse': rmse_ftr, 'mae' : mae_ftr},
'final': {'original': df['close'].tail(test_len).tolist(), 'prediction': final_prediction.values.tolist(), 'mse': mse,
'rmse': rmse, 'mae': mae },
'accuracy': {'prediction vs close': accuracy_1, 'prediction vs prediction': accuracy_2}}
# save simulation data here as checkpoint
with open('simulation2_data.json', 'w') as fp:
json.dump(simulation2, fp)
for ma in simulation2.keys():
print('\n' + ma)
print('Prediction vs Close:\t\t' + str(round(100*simulation2[ma]['accuracy']['prediction vs close'], 2))
+ '% Accuracy')
print('Prediction vs Prediction:\t' + str(round(100*simulation2[ma]['accuracy']['prediction vs prediction'], 2))
+ '% Accuracy')
print('MSE:\t', simulation2[ma]['final']['mse'],
'\nRMSE:\t', simulation2[ma]['final']['rmse'],
'\nMAPE:\t', simulation2[ma]['final']['mae'])#,
# '\nMAPE:\t', simulation[ma]['final']['mape'])
# else:
# break
elapsed = timeit.default_timer() - start_time
print('Runtime: mins:',elapsed/60)
SMA
SMA([input_arrays], [timeperiod=30])
Simple Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
17
Working on SMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.66 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4157.020, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3687.148, Time=0.07 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.27 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3458.651, Time=0.09 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3322.133, Time=0.13 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.01 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.08 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3324.133, Time=0.26 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 3.637 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1657.067
Date: Sun, 12 Dec 2021 AIC 3322.133
Time: 12:17:59 BIC 3340.897
Sample: 0 HQIC 3329.339
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1966 0.003 -387.226 0.000 -1.203 -1.191
ar.L2 -0.8952 0.006 -138.692 0.000 -0.908 -0.883
ar.L3 -0.3968 0.006 -68.284 0.000 -0.408 -0.385
sigma2 3.5858 0.017 214.535 0.000 3.553 3.619
===================================================================================
Ljung-Box (L1) (Q): 14.47 Jarque-Bera (JB): 2428881.42
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 271.99
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.03468, saving model to LSTM2.h5
81/81 - 7s - loss: 0.0868 - accuracy: 0.0000e+00 - val_loss: 0.0347 - val_accuracy: 0.0037 - lr: 0.0010 - 7s/epoch - 91ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.03468 to 0.01570, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0291 - accuracy: 0.0000e+00 - val_loss: 0.0157 - val_accuracy: 0.0037 - lr: 0.0010 - 736ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.01570
81/81 - 1s - loss: 0.0267 - accuracy: 0.0000e+00 - val_loss: 0.0366 - val_accuracy: 0.0037 - lr: 0.0010 - 701ms/epoch - 9ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.01570 to 0.00748, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0350 - accuracy: 0.0000e+00 - val_loss: 0.0075 - val_accuracy: 0.0037 - lr: 0.0010 - 757ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.00748 to 0.00470, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0181 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 0.0010 - 712ms/epoch - 9ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.00470 to 0.00433, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0171 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 0.0010 - 742ms/epoch - 9ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0184 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 0.0010 - 741ms/epoch - 9ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0190 - accuracy: 0.0000e+00 - val_loss: 0.0057 - val_accuracy: 0.0037 - lr: 0.0010 - 731ms/epoch - 9ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0197 - accuracy: 0.0000e+00 - val_loss: 0.0065 - val_accuracy: 0.0037 - lr: 0.0010 - 669ms/epoch - 8ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0198 - accuracy: 0.0000e+00 - val_loss: 0.0068 - val_accuracy: 0.0037 - lr: 0.0010 - 693ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00011: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0186 - accuracy: 0.0000e+00 - val_loss: 0.0065 - val_accuracy: 0.0037 - lr: 0.0010 - 708ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0234 - accuracy: 0.0000e+00 - val_loss: 0.0519 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 727ms/epoch - 9ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0046 - accuracy: 0.0000e+00 - val_loss: 0.0467 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 701ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0033 - accuracy: 0.0000e+00 - val_loss: 0.0426 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 687ms/epoch - 8ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0024 - accuracy: 0.0000e+00 - val_loss: 0.0384 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 694ms/epoch - 9ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00016: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0018 - accuracy: 0.0000e+00 - val_loss: 0.0341 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 709ms/epoch - 9ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0343 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 737ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0344 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 674ms/epoch - 8ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0343 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 686ms/epoch - 8ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0341 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 742ms/epoch - 9ms/step
Epoch 21/500
Epoch 00021: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00021: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0338 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 711ms/epoch - 9ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0335 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 685ms/epoch - 8ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0330 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 679ms/epoch - 8ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0326 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 690ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0320 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 717ms/epoch - 9ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0315 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 672ms/epoch - 8ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0310 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 747ms/epoch - 9ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0304 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 693ms/epoch - 9ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0298 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 683ms/epoch - 8ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0292 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 689ms/epoch - 9ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0286 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 666ms/epoch - 8ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0280 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 702ms/epoch - 9ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0273 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0267 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 734ms/epoch - 9ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0260 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 686ms/epoch - 8ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0254 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0247 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 690ms/epoch - 9ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0241 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 694ms/epoch - 9ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0234 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0227 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 698ms/epoch - 9ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00433
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0220 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.9353e-04 - accuracy: 0.0000e+00 - val_loss: 0.0214 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 700ms/epoch - 9ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.8646e-04 - accuracy: 0.0000e+00 - val_loss: 0.0207 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.7985e-04 - accuracy: 0.0000e+00 - val_loss: 0.0200 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 680ms/epoch - 8ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.7366e-04 - accuracy: 0.0000e+00 - val_loss: 0.0194 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.6786e-04 - accuracy: 0.0000e+00 - val_loss: 0.0187 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 703ms/epoch - 9ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.6241e-04 - accuracy: 0.0000e+00 - val_loss: 0.0181 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 689ms/epoch - 9ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.5728e-04 - accuracy: 0.0000e+00 - val_loss: 0.0175 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 699ms/epoch - 9ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.5243e-04 - accuracy: 0.0000e+00 - val_loss: 0.0169 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 733ms/epoch - 9ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.4784e-04 - accuracy: 0.0000e+00 - val_loss: 0.0162 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 692ms/epoch - 9ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.4348e-04 - accuracy: 0.0000e+00 - val_loss: 0.0156 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 717ms/epoch - 9ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.3932e-04 - accuracy: 0.0000e+00 - val_loss: 0.0151 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 731ms/epoch - 9ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.3533e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.3151e-04 - accuracy: 0.0000e+00 - val_loss: 0.0139 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.2781e-04 - accuracy: 0.0000e+00 - val_loss: 0.0134 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 711ms/epoch - 9ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00433
81/81 - 1s - loss: 9.2424e-04 - accuracy: 0.0000e+00 - val_loss: 0.0129 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 737ms/epoch - 9ms/step
Epoch 00056: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 136.8787627980392
RMSE: 11.699519767838302
MAPE: 9.782684450137419
EMA
EMA([input_arrays], [timeperiod=30])
Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
51
Working on EMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.57 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4231.556, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3761.238, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.41 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3532.227, Time=0.09 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3394.496, Time=0.13 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.18 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.91 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3396.496, Time=0.27 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 3.680 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1693.248
Date: Sun, 12 Dec 2021 AIC 3394.496
Time: 12:20:17 BIC 3413.260
Sample: 0 HQIC 3401.702
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1982 0.003 -389.569 0.000 -1.204 -1.192
ar.L2 -0.8976 0.006 -139.811 0.000 -0.910 -0.885
ar.L3 -0.3984 0.006 -68.662 0.000 -0.410 -0.387
sigma2 3.9230 0.018 215.372 0.000 3.887 3.959
===================================================================================
Ljung-Box (L1) (Q): 14.54 Jarque-Bera (JB): 2462173.05
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 273.82
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.06416, saving model to LSTM2.h5
81/81 - 6s - loss: 0.1950 - accuracy: 0.0000e+00 - val_loss: 0.0642 - val_accuracy: 0.0037 - lr: 0.0010 - 6s/epoch - 69ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.06416 to 0.01511, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0196 - accuracy: 0.0000e+00 - val_loss: 0.0151 - val_accuracy: 0.0037 - lr: 0.0010 - 702ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.01511
81/81 - 1s - loss: 0.0041 - accuracy: 0.0000e+00 - val_loss: 0.0225 - val_accuracy: 0.0037 - lr: 0.0010 - 719ms/epoch - 9ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.01511 to 0.01044, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0062 - accuracy: 0.0000e+00 - val_loss: 0.0104 - val_accuracy: 0.0037 - lr: 0.0010 - 739ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0111 - accuracy: 0.0000e+00 - val_loss: 0.0126 - val_accuracy: 0.0037 - lr: 0.0010 - 666ms/epoch - 8ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0177 - accuracy: 0.0000e+00 - val_loss: 0.0192 - val_accuracy: 0.0037 - lr: 0.0010 - 677ms/epoch - 8ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0174 - accuracy: 0.0000e+00 - val_loss: 0.0227 - val_accuracy: 0.0037 - lr: 0.0010 - 656ms/epoch - 8ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0131 - accuracy: 0.0000e+00 - val_loss: 0.0333 - val_accuracy: 0.0037 - lr: 0.0010 - 676ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00009: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0082 - accuracy: 0.0000e+00 - val_loss: 0.0487 - val_accuracy: 0.0037 - lr: 0.0010 - 676ms/epoch - 8ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0130 - accuracy: 0.0000e+00 - val_loss: 0.0795 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 677ms/epoch - 8ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0093 - accuracy: 0.0000e+00 - val_loss: 0.0790 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 697ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0082 - accuracy: 0.0000e+00 - val_loss: 0.0843 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 676ms/epoch - 8ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0073 - accuracy: 0.0000e+00 - val_loss: 0.0910 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 677ms/epoch - 8ms/step
Epoch 14/500
Epoch 00014: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00014: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0066 - accuracy: 0.0000e+00 - val_loss: 0.0974 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 686ms/epoch - 8ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0060 - accuracy: 0.0000e+00 - val_loss: 0.1040 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 684ms/epoch - 8ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0053 - accuracy: 0.0000e+00 - val_loss: 0.1093 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 683ms/epoch - 8ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0049 - accuracy: 0.0000e+00 - val_loss: 0.1133 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 684ms/epoch - 8ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0046 - accuracy: 0.0000e+00 - val_loss: 0.1163 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 675ms/epoch - 8ms/step
Epoch 19/500
Epoch 00019: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00019: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0045 - accuracy: 0.0000e+00 - val_loss: 0.1185 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 675ms/epoch - 8ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0044 - accuracy: 0.0000e+00 - val_loss: 0.1201 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 684ms/epoch - 8ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0043 - accuracy: 0.0000e+00 - val_loss: 0.1213 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 681ms/epoch - 8ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0042 - accuracy: 0.0000e+00 - val_loss: 0.1222 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 671ms/epoch - 8ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0042 - accuracy: 0.0000e+00 - val_loss: 0.1230 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 697ms/epoch - 9ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0041 - accuracy: 0.0000e+00 - val_loss: 0.1237 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 731ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0041 - accuracy: 0.0000e+00 - val_loss: 0.1243 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 676ms/epoch - 8ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0040 - accuracy: 0.0000e+00 - val_loss: 0.1249 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 697ms/epoch - 9ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0040 - accuracy: 0.0000e+00 - val_loss: 0.1256 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 689ms/epoch - 9ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0039 - accuracy: 0.0000e+00 - val_loss: 0.1262 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 670ms/epoch - 8ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0039 - accuracy: 0.0000e+00 - val_loss: 0.1268 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 705ms/epoch - 9ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0038 - accuracy: 0.0000e+00 - val_loss: 0.1275 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 724ms/epoch - 9ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0037 - accuracy: 0.0000e+00 - val_loss: 0.1281 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 697ms/epoch - 9ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0037 - accuracy: 0.0000e+00 - val_loss: 0.1288 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 673ms/epoch - 8ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0036 - accuracy: 0.0000e+00 - val_loss: 0.1295 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 676ms/epoch - 8ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0036 - accuracy: 0.0000e+00 - val_loss: 0.1301 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 693ms/epoch - 9ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0035 - accuracy: 0.0000e+00 - val_loss: 0.1307 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 681ms/epoch - 8ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0035 - accuracy: 0.0000e+00 - val_loss: 0.1313 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0034 - accuracy: 0.0000e+00 - val_loss: 0.1318 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 701ms/epoch - 9ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0034 - accuracy: 0.0000e+00 - val_loss: 0.1324 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 672ms/epoch - 8ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0033 - accuracy: 0.0000e+00 - val_loss: 0.1328 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 686ms/epoch - 8ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0033 - accuracy: 0.0000e+00 - val_loss: 0.1332 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 687ms/epoch - 8ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0032 - accuracy: 0.0000e+00 - val_loss: 0.1335 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 678ms/epoch - 8ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0032 - accuracy: 0.0000e+00 - val_loss: 0.1338 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 693ms/epoch - 9ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0031 - accuracy: 0.0000e+00 - val_loss: 0.1340 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 662ms/epoch - 8ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0031 - accuracy: 0.0000e+00 - val_loss: 0.1341 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 672ms/epoch - 8ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0030 - accuracy: 0.0000e+00 - val_loss: 0.1341 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 672ms/epoch - 8ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0030 - accuracy: 0.0000e+00 - val_loss: 0.1341 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 682ms/epoch - 8ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0029 - accuracy: 0.0000e+00 - val_loss: 0.1339 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 667ms/epoch - 8ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0029 - accuracy: 0.0000e+00 - val_loss: 0.1337 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 729ms/epoch - 9ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0028 - accuracy: 0.0000e+00 - val_loss: 0.1333 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0028 - accuracy: 0.0000e+00 - val_loss: 0.1329 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 675ms/epoch - 8ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0027 - accuracy: 0.0000e+00 - val_loss: 0.1323 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0027 - accuracy: 0.0000e+00 - val_loss: 0.1317 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0027 - accuracy: 0.0000e+00 - val_loss: 0.1309 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 693ms/epoch - 9ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.01044
81/81 - 1s - loss: 0.0026 - accuracy: 0.0000e+00 - val_loss: 0.1301 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 679ms/epoch - 8ms/step
Epoch 00054: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 136.8787627980392
RMSE: 11.699519767838302
MAPE: 9.782684450137419
EMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 44.78% Accuracy
MSE: 325.2876788257106
RMSE: 18.035733387520192
MAPE: 15.58326491461623
WMA
WMA([input_arrays], [timeperiod=30])
Weighted Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
49
Working on WMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.57 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4264.089, Time=0.05 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3793.930, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.34 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3564.923, Time=0.09 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3427.258, Time=0.12 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.68 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.62 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3429.258, Time=0.27 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 3.816 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1709.629
Date: Sun, 12 Dec 2021 AIC 3427.258
Time: 12:22:33 BIC 3446.021
Sample: 0 HQIC 3434.464
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1981 0.003 -389.386 0.000 -1.204 -1.192
ar.L2 -0.8974 0.006 -139.699 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.737 0.000 -0.410 -0.387
sigma2 4.0860 0.019 215.311 0.000 4.049 4.123
===================================================================================
Ljung-Box (L1) (Q): 14.57 Jarque-Bera (JB): 2460901.70
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 273.75
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.17237, saving model to LSTM2.h5
81/81 - 6s - loss: 0.1566 - accuracy: 0.0000e+00 - val_loss: 0.1724 - val_accuracy: 0.0000e+00 - lr: 0.0010 - 6s/epoch - 76ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.17237 to 0.01302, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0208 - accuracy: 0.0000e+00 - val_loss: 0.0130 - val_accuracy: 0.0037 - lr: 0.0010 - 744ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.01302
81/81 - 1s - loss: 0.0153 - accuracy: 0.0000e+00 - val_loss: 0.0210 - val_accuracy: 0.0037 - lr: 0.0010 - 686ms/epoch - 8ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.01302 to 0.00607, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0182 - accuracy: 0.0000e+00 - val_loss: 0.0061 - val_accuracy: 0.0037 - lr: 0.0010 - 709ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.00607 to 0.00465, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0144 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 0.0010 - 725ms/epoch - 9ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0181 - accuracy: 0.0000e+00 - val_loss: 0.0048 - val_accuracy: 0.0037 - lr: 0.0010 - 726ms/epoch - 9ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0242 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 0.0010 - 724ms/epoch - 9ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0260 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 0.0010 - 683ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0239 - accuracy: 0.0000e+00 - val_loss: 0.0052 - val_accuracy: 0.0037 - lr: 0.0010 - 692ms/epoch - 9ms/step
Epoch 10/500
Epoch 00010: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00010: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0197 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 0.0010 - 689ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0184 - accuracy: 0.0000e+00 - val_loss: 0.0307 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 679ms/epoch - 8ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0041 - accuracy: 0.0000e+00 - val_loss: 0.0268 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 674ms/epoch - 8ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0029 - accuracy: 0.0000e+00 - val_loss: 0.0233 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 703ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0022 - accuracy: 0.0000e+00 - val_loss: 0.0203 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 704ms/epoch - 9ms/step
Epoch 15/500
Epoch 00015: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00015: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0017 - accuracy: 0.0000e+00 - val_loss: 0.0177 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 695ms/epoch - 9ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0177 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 705ms/epoch - 9ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0177 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 690ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0176 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 672ms/epoch - 8ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0174 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 690ms/epoch - 9ms/step
Epoch 20/500
Epoch 00020: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00020: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0173 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 688ms/epoch - 8ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0170 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 672ms/epoch - 8ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0168 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 733ms/epoch - 9ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0165 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 739ms/epoch - 9ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0163 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 684ms/epoch - 8ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0160 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 680ms/epoch - 8ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0157 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 680ms/epoch - 8ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0154 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 673ms/epoch - 8ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0151 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 664ms/epoch - 8ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0148 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 663ms/epoch - 8ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 741ms/epoch - 9ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0141 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 703ms/epoch - 9ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0138 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 732ms/epoch - 9ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0135 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 744ms/epoch - 9ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0132 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 688ms/epoch - 8ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0129 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 665ms/epoch - 8ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0126 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 711ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0122 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 688ms/epoch - 8ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0119 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 682ms/epoch - 8ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0116 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 683ms/epoch - 8ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0113 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 739ms/epoch - 9ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0110 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0107 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 687ms/epoch - 8ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0104 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 679ms/epoch - 8ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0101 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 679ms/epoch - 8ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0098 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 702ms/epoch - 9ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0095 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 666ms/epoch - 8ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0092 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 708ms/epoch - 9ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0090 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 735ms/epoch - 9ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00465
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0087 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 676ms/epoch - 8ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00465
81/81 - 1s - loss: 9.9766e-04 - accuracy: 0.0000e+00 - val_loss: 0.0084 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 729ms/epoch - 9ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00465
81/81 - 1s - loss: 9.9414e-04 - accuracy: 0.0000e+00 - val_loss: 0.0082 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 681ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00465
81/81 - 1s - loss: 9.9079e-04 - accuracy: 0.0000e+00 - val_loss: 0.0079 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00465
81/81 - 1s - loss: 9.8760e-04 - accuracy: 0.0000e+00 - val_loss: 0.0077 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 722ms/epoch - 9ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00465
81/81 - 1s - loss: 9.8453e-04 - accuracy: 0.0000e+00 - val_loss: 0.0075 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 701ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00465
81/81 - 1s - loss: 9.8158e-04 - accuracy: 0.0000e+00 - val_loss: 0.0072 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 697ms/epoch - 9ms/step
Epoch 00055: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 136.8787627980392
RMSE: 11.699519767838302
MAPE: 9.782684450137419
EMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 44.78% Accuracy
MSE: 325.2876788257106
RMSE: 18.035733387520192
MAPE: 15.58326491461623
WMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.28% Accuracy
MSE: 142.26654445774975
RMSE: 11.927554001460221
MAPE: 9.787176512242624
DEMA
DEMA([input_arrays], [timeperiod=30])
Double Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
89
Working on DEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.58 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4436.126, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3965.317, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.53 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3736.589, Time=0.08 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3598.951, Time=0.11 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.25 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.29 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3600.951, Time=0.24 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.207 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1795.475
Date: Sun, 12 Dec 2021 AIC 3598.951
Time: 12:24:49 BIC 3617.714
Sample: 0 HQIC 3606.157
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1983 0.003 -389.581 0.000 -1.204 -1.192
ar.L2 -0.8973 0.006 -139.732 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.649 0.000 -0.410 -0.387
sigma2 5.0573 0.023 215.292 0.000 5.011 5.103
===================================================================================
Ljung-Box (L1) (Q): 14.41 Jarque-Bera (JB): 2460553.80
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.89
Prob(H) (two-sided): 0.00 Kurtosis: 273.74
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.07371, saving model to LSTM2.h5
81/81 - 6s - loss: 0.0925 - accuracy: 0.0000e+00 - val_loss: 0.0737 - val_accuracy: 0.0037 - lr: 0.0010 - 6s/epoch - 77ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.07371 to 0.01875, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0304 - accuracy: 0.0000e+00 - val_loss: 0.0187 - val_accuracy: 0.0037 - lr: 0.0010 - 735ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.01875 to 0.00901, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0406 - accuracy: 0.0000e+00 - val_loss: 0.0090 - val_accuracy: 0.0037 - lr: 0.0010 - 738ms/epoch - 9ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.00901 to 0.00558, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0309 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 0.0010 - 728ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.00558 to 0.00430, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0301 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 0.0010 - 746ms/epoch - 9ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.00430 to 0.00422, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0165 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 0.0010 - 728ms/epoch - 9ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.00422 to 0.00343, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0108 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 0.0010 - 736ms/epoch - 9ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.00343 to 0.00311, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0088 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 0.0010 - 743ms/epoch - 9ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00311
81/81 - 1s - loss: 0.0087 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 0.0010 - 709ms/epoch - 9ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00311
81/81 - 1s - loss: 0.0096 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 0.0010 - 699ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.00311
81/81 - 1s - loss: 0.0110 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 0.0010 - 693ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00311
81/81 - 1s - loss: 0.0122 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 0.0010 - 708ms/epoch - 9ms/step
Epoch 13/500
Epoch 00013: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00013: val_loss did not improve from 0.00311
81/81 - 1s - loss: 0.0128 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 0.0010 - 707ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00311
81/81 - 1s - loss: 0.0172 - accuracy: 0.0000e+00 - val_loss: 0.0092 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 737ms/epoch - 9ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00311
81/81 - 1s - loss: 0.0035 - accuracy: 0.0000e+00 - val_loss: 0.0063 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 725ms/epoch - 9ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00311
81/81 - 1s - loss: 0.0025 - accuracy: 0.0000e+00 - val_loss: 0.0048 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 689ms/epoch - 9ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00311
81/81 - 1s - loss: 0.0019 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 698ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss improved from 0.00311 to 0.00292, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0015 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 905ms/epoch - 11ms/step
Epoch 19/500
Epoch 00019: val_loss improved from 0.00292 to 0.00250, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 735ms/epoch - 9ms/step
Epoch 20/500
Epoch 00020: val_loss improved from 0.00250 to 0.00242, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0024 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 744ms/epoch - 9ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00242
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 690ms/epoch - 9ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00242
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 715ms/epoch - 9ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00242
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 714ms/epoch - 9ms/step
Epoch 24/500
Epoch 00024: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00024: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.9079e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 691ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.8979e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 721ms/epoch - 9ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.5139e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 709ms/epoch - 9ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.3537e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 715ms/epoch - 9ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.2832e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 694ms/epoch - 9ms/step
Epoch 29/500
Epoch 00029: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00029: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.2482e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 721ms/epoch - 9ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.2276e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 709ms/epoch - 9ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.2131e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.2015e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 728ms/epoch - 9ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1911e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 720ms/epoch - 9ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1814e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 701ms/epoch - 9ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1719e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 708ms/epoch - 9ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1625e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 755ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1531e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1435e-04 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 705ms/epoch - 9ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1338e-04 - accuracy: 0.0000e+00 - val_loss: 0.0048 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 709ms/epoch - 9ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1240e-04 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 703ms/epoch - 9ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1140e-04 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 706ms/epoch - 9ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.1038e-04 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.0935e-04 - accuracy: 0.0000e+00 - val_loss: 0.0051 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 701ms/epoch - 9ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.0829e-04 - accuracy: 0.0000e+00 - val_loss: 0.0052 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 700ms/epoch - 9ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.0722e-04 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 694ms/epoch - 9ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.0612e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.0500e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 704ms/epoch - 9ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.0385e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 673ms/epoch - 8ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.0269e-04 - accuracy: 0.0000e+00 - val_loss: 0.0058 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 693ms/epoch - 9ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.0149e-04 - accuracy: 0.0000e+00 - val_loss: 0.0059 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 704ms/epoch - 9ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00242
81/81 - 1s - loss: 9.0027e-04 - accuracy: 0.0000e+00 - val_loss: 0.0060 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 684ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.9903e-04 - accuracy: 0.0000e+00 - val_loss: 0.0061 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 733ms/epoch - 9ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.9775e-04 - accuracy: 0.0000e+00 - val_loss: 0.0063 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 711ms/epoch - 9ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.9645e-04 - accuracy: 0.0000e+00 - val_loss: 0.0064 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 736ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.9512e-04 - accuracy: 0.0000e+00 - val_loss: 0.0066 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 730ms/epoch - 9ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.9376e-04 - accuracy: 0.0000e+00 - val_loss: 0.0067 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.9238e-04 - accuracy: 0.0000e+00 - val_loss: 0.0068 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 704ms/epoch - 9ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.9096e-04 - accuracy: 0.0000e+00 - val_loss: 0.0070 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 688ms/epoch - 8ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.8951e-04 - accuracy: 0.0000e+00 - val_loss: 0.0072 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 686ms/epoch - 8ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.8803e-04 - accuracy: 0.0000e+00 - val_loss: 0.0073 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 710ms/epoch - 9ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.8652e-04 - accuracy: 0.0000e+00 - val_loss: 0.0075 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 680ms/epoch - 8ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.8497e-04 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 679ms/epoch - 8ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.8340e-04 - accuracy: 0.0000e+00 - val_loss: 0.0078 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 664ms/epoch - 8ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.8179e-04 - accuracy: 0.0000e+00 - val_loss: 0.0080 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 674ms/epoch - 8ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.8015e-04 - accuracy: 0.0000e+00 - val_loss: 0.0082 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 669ms/epoch - 8ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.7847e-04 - accuracy: 0.0000e+00 - val_loss: 0.0083 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 668ms/epoch - 8ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.7676e-04 - accuracy: 0.0000e+00 - val_loss: 0.0085 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 658ms/epoch - 8ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.7502e-04 - accuracy: 0.0000e+00 - val_loss: 0.0087 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.7325e-04 - accuracy: 0.0000e+00 - val_loss: 0.0089 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 660ms/epoch - 8ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.00242
81/81 - 1s - loss: 8.7144e-04 - accuracy: 0.0000e+00 - val_loss: 0.0090 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 692ms/epoch - 9ms/step
Epoch 00070: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 136.8787627980392
RMSE: 11.699519767838302
MAPE: 9.782684450137419
EMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 44.78% Accuracy
MSE: 325.2876788257106
RMSE: 18.035733387520192
MAPE: 15.58326491461623
WMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.28% Accuracy
MSE: 142.26654445774975
RMSE: 11.927554001460221
MAPE: 9.787176512242624
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 171.26734938505615
RMSE: 13.086915197442679
MAPE: 11.821213958102536
KAMA
KAMA([input_arrays], [timeperiod=30])
Kaufman Adaptive Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
18
Working on KAMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.53 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4190.464, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3724.371, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.39 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3494.154, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3357.435, Time=0.13 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.57 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.99 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3359.435, Time=0.28 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.087 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1674.717
Date: Sun, 12 Dec 2021 AIC 3357.435
Time: 12:27:40 BIC 3376.198
Sample: 0 HQIC 3364.641
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1955 0.003 -381.246 0.000 -1.202 -1.189
ar.L2 -0.8964 0.007 -135.835 0.000 -0.909 -0.883
ar.L3 -0.3971 0.006 -67.229 0.000 -0.409 -0.385
sigma2 3.7466 0.018 211.623 0.000 3.712 3.781
===================================================================================
Ljung-Box (L1) (Q): 14.20 Jarque-Bera (JB): 2338363.32
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.01 Skew: 3.76
Prob(H) (two-sided): 0.00 Kurtosis: 266.93
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.09321, saving model to LSTM2.h5
81/81 - 6s - loss: 0.1163 - accuracy: 0.0000e+00 - val_loss: 0.0932 - val_accuracy: 0.0037 - lr: 0.0010 - 6s/epoch - 71ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.09321 to 0.07989, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0552 - accuracy: 0.0000e+00 - val_loss: 0.0799 - val_accuracy: 0.0037 - lr: 0.0010 - 750ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.07989 to 0.01022, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0555 - accuracy: 0.0000e+00 - val_loss: 0.0102 - val_accuracy: 0.0037 - lr: 0.0010 - 741ms/epoch - 9ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.01022 to 0.00514, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0274 - accuracy: 0.0000e+00 - val_loss: 0.0051 - val_accuracy: 0.0037 - lr: 0.0010 - 702ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.00514
81/81 - 1s - loss: 0.0227 - accuracy: 0.0000e+00 - val_loss: 0.0059 - val_accuracy: 0.0037 - lr: 0.0010 - 707ms/epoch - 9ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.00514 to 0.00433, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0094 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 0.0010 - 762ms/epoch - 9ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.00433 to 0.00403, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0060 - accuracy: 0.0000e+00 - val_loss: 0.0040 - val_accuracy: 0.0037 - lr: 0.0010 - 717ms/epoch - 9ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.00403 to 0.00382, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0054 - accuracy: 0.0000e+00 - val_loss: 0.0038 - val_accuracy: 0.0037 - lr: 0.0010 - 729ms/epoch - 9ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.00382 to 0.00375, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0056 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 0.0010 - 741ms/epoch - 9ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00375
81/81 - 1s - loss: 0.0078 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 0.0010 - 705ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.00375
81/81 - 1s - loss: 0.0100 - accuracy: 0.0000e+00 - val_loss: 0.0051 - val_accuracy: 0.0037 - lr: 0.0010 - 700ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00375
81/81 - 1s - loss: 0.0148 - accuracy: 0.0000e+00 - val_loss: 0.0052 - val_accuracy: 0.0037 - lr: 0.0010 - 749ms/epoch - 9ms/step
Epoch 13/500
Epoch 00013: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00013: val_loss did not improve from 0.00375
81/81 - 1s - loss: 0.0163 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 0.0010 - 691ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00375
81/81 - 1s - loss: 0.0203 - accuracy: 0.0000e+00 - val_loss: 0.0074 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 673ms/epoch - 8ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00375
81/81 - 1s - loss: 0.0051 - accuracy: 0.0000e+00 - val_loss: 0.0060 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 700ms/epoch - 9ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00375
81/81 - 1s - loss: 0.0035 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 693ms/epoch - 9ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00375
81/81 - 1s - loss: 0.0025 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 689ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss improved from 0.00375 to 0.00307, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0019 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 795ms/epoch - 10ms/step
Epoch 19/500
Epoch 00019: val_loss improved from 0.00307 to 0.00258, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0015 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 737ms/epoch - 9ms/step
Epoch 20/500
Epoch 00020: val_loss improved from 0.00258 to 0.00241, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0024 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 725ms/epoch - 9ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00241
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 699ms/epoch - 9ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00241
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 714ms/epoch - 9ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00241
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 710ms/epoch - 9ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00241
81/81 - 1s - loss: 9.6716e-04 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 704ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00025: val_loss did not improve from 0.00241
81/81 - 1s - loss: 9.4434e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 714ms/epoch - 9ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00241
81/81 - 1s - loss: 9.1698e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 698ms/epoch - 9ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00241
81/81 - 1s - loss: 9.0828e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 724ms/epoch - 9ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00241
81/81 - 1s - loss: 9.0429e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 737ms/epoch - 9ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00241
81/81 - 1s - loss: 9.0209e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 739ms/epoch - 9ms/step
Epoch 30/500
Epoch 00030: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00030: val_loss did not improve from 0.00241
81/81 - 1s - loss: 9.0056e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 708ms/epoch - 9ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.9929e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 661ms/epoch - 8ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.9813e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 676ms/epoch - 8ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.9702e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 660ms/epoch - 8ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.9592e-04 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 665ms/epoch - 8ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.9481e-04 - accuracy: 0.0000e+00 - val_loss: 0.0048 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 674ms/epoch - 8ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.9370e-04 - accuracy: 0.0000e+00 - val_loss: 0.0048 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 731ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.9258e-04 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 676ms/epoch - 8ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.9145e-04 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 679ms/epoch - 8ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.9030e-04 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 681ms/epoch - 8ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.8913e-04 - accuracy: 0.0000e+00 - val_loss: 0.0051 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 675ms/epoch - 8ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.8795e-04 - accuracy: 0.0000e+00 - val_loss: 0.0051 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 688ms/epoch - 8ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.8676e-04 - accuracy: 0.0000e+00 - val_loss: 0.0052 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 670ms/epoch - 8ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.8554e-04 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 707ms/epoch - 9ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.8431e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 682ms/epoch - 8ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.8307e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 666ms/epoch - 8ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.8181e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 667ms/epoch - 8ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.8053e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 683ms/epoch - 8ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.7923e-04 - accuracy: 0.0000e+00 - val_loss: 0.0057 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 676ms/epoch - 8ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.7792e-04 - accuracy: 0.0000e+00 - val_loss: 0.0058 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 661ms/epoch - 8ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.7659e-04 - accuracy: 0.0000e+00 - val_loss: 0.0059 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 742ms/epoch - 9ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.7523e-04 - accuracy: 0.0000e+00 - val_loss: 0.0060 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 680ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.7387e-04 - accuracy: 0.0000e+00 - val_loss: 0.0061 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 681ms/epoch - 8ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.7248e-04 - accuracy: 0.0000e+00 - val_loss: 0.0062 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 689ms/epoch - 9ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.7108e-04 - accuracy: 0.0000e+00 - val_loss: 0.0063 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 690ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.6965e-04 - accuracy: 0.0000e+00 - val_loss: 0.0065 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 678ms/epoch - 8ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.6821e-04 - accuracy: 0.0000e+00 - val_loss: 0.0066 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 662ms/epoch - 8ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.6675e-04 - accuracy: 0.0000e+00 - val_loss: 0.0067 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 662ms/epoch - 8ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.6526e-04 - accuracy: 0.0000e+00 - val_loss: 0.0068 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 677ms/epoch - 8ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.6376e-04 - accuracy: 0.0000e+00 - val_loss: 0.0069 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 683ms/epoch - 8ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.6224e-04 - accuracy: 0.0000e+00 - val_loss: 0.0070 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 687ms/epoch - 8ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.6070e-04 - accuracy: 0.0000e+00 - val_loss: 0.0072 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 682ms/epoch - 8ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.5914e-04 - accuracy: 0.0000e+00 - val_loss: 0.0073 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 677ms/epoch - 8ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.5756e-04 - accuracy: 0.0000e+00 - val_loss: 0.0074 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 687ms/epoch - 8ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.5596e-04 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 732ms/epoch - 9ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.5434e-04 - accuracy: 0.0000e+00 - val_loss: 0.0077 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 699ms/epoch - 9ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.5269e-04 - accuracy: 0.0000e+00 - val_loss: 0.0078 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 746ms/epoch - 9ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.5103e-04 - accuracy: 0.0000e+00 - val_loss: 0.0079 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 685ms/epoch - 8ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.4934e-04 - accuracy: 0.0000e+00 - val_loss: 0.0081 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 680ms/epoch - 8ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.4764e-04 - accuracy: 0.0000e+00 - val_loss: 0.0082 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 735ms/epoch - 9ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.00241
81/81 - 1s - loss: 8.4591e-04 - accuracy: 0.0000e+00 - val_loss: 0.0083 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 728ms/epoch - 9ms/step
Epoch 00070: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 136.8787627980392
RMSE: 11.699519767838302
MAPE: 9.782684450137419
EMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 44.78% Accuracy
MSE: 325.2876788257106
RMSE: 18.035733387520192
MAPE: 15.58326491461623
WMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.28% Accuracy
MSE: 142.26654445774975
RMSE: 11.927554001460221
MAPE: 9.787176512242624
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 171.26734938505615
RMSE: 13.086915197442679
MAPE: 11.821213958102536
KAMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 52.24625253706662
RMSE: 7.228156925321048
MAPE: 5.787764162926424
MIDPOINT
MIDPOINT([input_arrays], [timeperiod=14])
MidPoint over period (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 14
Outputs:
real
14
Working on MIDPOINT predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.50 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4212.289, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3747.746, Time=0.07 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.33 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3523.401, Time=0.11 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3387.759, Time=0.12 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.61 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.18 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3389.758, Time=0.30 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.270 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1689.879
Date: Sun, 12 Dec 2021 AIC 3387.759
Time: 12:30:07 BIC 3406.522
Sample: 0 HQIC 3394.964
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1878 0.003 -345.315 0.000 -1.195 -1.181
ar.L2 -0.8876 0.007 -121.809 0.000 -0.902 -0.873
ar.L3 -0.3957 0.007 -60.127 0.000 -0.409 -0.383
sigma2 3.8904 0.020 193.404 0.000 3.851 3.930
===================================================================================
Ljung-Box (L1) (Q): 13.21 Jarque-Bera (JB): 1659080.01
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.08 Skew: 3.28
Prob(H) (two-sided): 0.00 Kurtosis: 225.31
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.07124, saving model to LSTM2.h5
81/81 - 6s - loss: 0.1063 - accuracy: 0.0000e+00 - val_loss: 0.0712 - val_accuracy: 0.0037 - lr: 0.0010 - 6s/epoch - 74ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.07124 to 0.04279, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0388 - accuracy: 0.0000e+00 - val_loss: 0.0428 - val_accuracy: 0.0037 - lr: 0.0010 - 748ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.04279 to 0.01224, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0581 - accuracy: 0.0000e+00 - val_loss: 0.0122 - val_accuracy: 0.0037 - lr: 0.0010 - 747ms/epoch - 9ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.01224 to 0.00545, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0288 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 0.0010 - 727ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.00545 to 0.00465, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0259 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 0.0010 - 738ms/epoch - 9ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.00465 to 0.00417, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0109 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 0.0010 - 739ms/epoch - 9ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.00417
81/81 - 1s - loss: 0.0063 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 0.0010 - 696ms/epoch - 9ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.00417
81/81 - 1s - loss: 0.0050 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 0.0010 - 671ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00417
81/81 - 1s - loss: 0.0053 - accuracy: 0.0000e+00 - val_loss: 0.0064 - val_accuracy: 0.0037 - lr: 0.0010 - 698ms/epoch - 9ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00417
81/81 - 1s - loss: 0.0066 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 0.0010 - 702ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00011: val_loss did not improve from 0.00417
81/81 - 1s - loss: 0.0089 - accuracy: 0.0000e+00 - val_loss: 0.0085 - val_accuracy: 0.0037 - lr: 0.0010 - 723ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00417
81/81 - 1s - loss: 0.0169 - accuracy: 0.0000e+00 - val_loss: 0.0060 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 694ms/epoch - 9ms/step
Epoch 13/500
Epoch 00013: val_loss improved from 0.00417 to 0.00384, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0041 - accuracy: 0.0000e+00 - val_loss: 0.0038 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 755ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss improved from 0.00384 to 0.00340, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0029 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 728ms/epoch - 9ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00340
81/81 - 1s - loss: 0.0021 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 760ms/epoch - 9ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00340
81/81 - 1s - loss: 0.0016 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 695ms/epoch - 9ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00340
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0052 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 702ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.00340
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0061 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 691ms/epoch - 9ms/step
Epoch 19/500
Epoch 00019: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00019: val_loss did not improve from 0.00340
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0070 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 702ms/epoch - 9ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.6145e-04 - accuracy: 0.0000e+00 - val_loss: 0.0072 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 672ms/epoch - 8ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.5517e-04 - accuracy: 0.0000e+00 - val_loss: 0.0073 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 692ms/epoch - 9ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.5095e-04 - accuracy: 0.0000e+00 - val_loss: 0.0074 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 748ms/epoch - 9ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.4738e-04 - accuracy: 0.0000e+00 - val_loss: 0.0074 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 717ms/epoch - 9ms/step
Epoch 24/500
Epoch 00024: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00024: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.4399e-04 - accuracy: 0.0000e+00 - val_loss: 0.0075 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 698ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.4064e-04 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 720ms/epoch - 9ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.3730e-04 - accuracy: 0.0000e+00 - val_loss: 0.0077 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 749ms/epoch - 9ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.3396e-04 - accuracy: 0.0000e+00 - val_loss: 0.0078 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 674ms/epoch - 8ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.3063e-04 - accuracy: 0.0000e+00 - val_loss: 0.0079 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 708ms/epoch - 9ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.2732e-04 - accuracy: 0.0000e+00 - val_loss: 0.0080 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 728ms/epoch - 9ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.2402e-04 - accuracy: 0.0000e+00 - val_loss: 0.0080 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 702ms/epoch - 9ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.2074e-04 - accuracy: 0.0000e+00 - val_loss: 0.0081 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 755ms/epoch - 9ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.1749e-04 - accuracy: 0.0000e+00 - val_loss: 0.0082 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 705ms/epoch - 9ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.1427e-04 - accuracy: 0.0000e+00 - val_loss: 0.0083 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 716ms/epoch - 9ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.1108e-04 - accuracy: 0.0000e+00 - val_loss: 0.0084 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 704ms/epoch - 9ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.0793e-04 - accuracy: 0.0000e+00 - val_loss: 0.0086 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 726ms/epoch - 9ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.0480e-04 - accuracy: 0.0000e+00 - val_loss: 0.0087 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 722ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00340
81/81 - 1s - loss: 9.0171e-04 - accuracy: 0.0000e+00 - val_loss: 0.0088 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.9866e-04 - accuracy: 0.0000e+00 - val_loss: 0.0089 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 682ms/epoch - 8ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.9564e-04 - accuracy: 0.0000e+00 - val_loss: 0.0090 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 727ms/epoch - 9ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.9266e-04 - accuracy: 0.0000e+00 - val_loss: 0.0091 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 700ms/epoch - 9ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.8971e-04 - accuracy: 0.0000e+00 - val_loss: 0.0092 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 718ms/epoch - 9ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.8680e-04 - accuracy: 0.0000e+00 - val_loss: 0.0093 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 732ms/epoch - 9ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.8393e-04 - accuracy: 0.0000e+00 - val_loss: 0.0094 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 697ms/epoch - 9ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.8108e-04 - accuracy: 0.0000e+00 - val_loss: 0.0096 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 730ms/epoch - 9ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.7827e-04 - accuracy: 0.0000e+00 - val_loss: 0.0097 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 757ms/epoch - 9ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.7549e-04 - accuracy: 0.0000e+00 - val_loss: 0.0098 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.7274e-04 - accuracy: 0.0000e+00 - val_loss: 0.0099 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 705ms/epoch - 9ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.7003e-04 - accuracy: 0.0000e+00 - val_loss: 0.0100 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 705ms/epoch - 9ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.6734e-04 - accuracy: 0.0000e+00 - val_loss: 0.0101 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 694ms/epoch - 9ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.6467e-04 - accuracy: 0.0000e+00 - val_loss: 0.0102 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 712ms/epoch - 9ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.6204e-04 - accuracy: 0.0000e+00 - val_loss: 0.0103 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.5942e-04 - accuracy: 0.0000e+00 - val_loss: 0.0104 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 702ms/epoch - 9ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.5684e-04 - accuracy: 0.0000e+00 - val_loss: 0.0105 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 720ms/epoch - 9ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.5427e-04 - accuracy: 0.0000e+00 - val_loss: 0.0106 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 707ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.5172e-04 - accuracy: 0.0000e+00 - val_loss: 0.0107 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 692ms/epoch - 9ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.4920e-04 - accuracy: 0.0000e+00 - val_loss: 0.0108 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.4669e-04 - accuracy: 0.0000e+00 - val_loss: 0.0109 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 724ms/epoch - 9ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.4421e-04 - accuracy: 0.0000e+00 - val_loss: 0.0109 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 719ms/epoch - 9ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.4174e-04 - accuracy: 0.0000e+00 - val_loss: 0.0110 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.3928e-04 - accuracy: 0.0000e+00 - val_loss: 0.0111 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 706ms/epoch - 9ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.3684e-04 - accuracy: 0.0000e+00 - val_loss: 0.0111 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 688ms/epoch - 8ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.3442e-04 - accuracy: 0.0000e+00 - val_loss: 0.0112 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 688ms/epoch - 8ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.3201e-04 - accuracy: 0.0000e+00 - val_loss: 0.0113 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 700ms/epoch - 9ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.00340
81/81 - 1s - loss: 8.2960e-04 - accuracy: 0.0000e+00 - val_loss: 0.0113 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 700ms/epoch - 9ms/step
Epoch 00064: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 136.8787627980392
RMSE: 11.699519767838302
MAPE: 9.782684450137419
EMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 44.78% Accuracy
MSE: 325.2876788257106
RMSE: 18.035733387520192
MAPE: 15.58326491461623
WMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.28% Accuracy
MSE: 142.26654445774975
RMSE: 11.927554001460221
MAPE: 9.787176512242624
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 171.26734938505615
RMSE: 13.086915197442679
MAPE: 11.821213958102536
KAMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 52.24625253706662
RMSE: 7.228156925321048
MAPE: 5.787764162926424
MIDPOINT
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 51.03620830548878
RMSE: 7.143963067197981
MAPE: 5.7689786092745114
T3
T3([input_arrays], [timeperiod=5], [vfactor=0.7])
Triple Exponential Moving Average (T3) (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 5
vfactor: 0.7
Outputs:
real
19
Working on T3 predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.54 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4414.515, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3944.062, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.51 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3715.173, Time=0.09 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3577.471, Time=0.11 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.86 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.81 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3579.471, Time=0.25 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.258 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1784.736
Date: Sun, 12 Dec 2021 AIC 3577.471
Time: 12:32:30 BIC 3596.235
Sample: 0 HQIC 3584.677
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1982 0.003 -389.844 0.000 -1.204 -1.192
ar.L2 -0.8974 0.006 -139.861 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.862 0.000 -0.410 -0.387
sigma2 4.9242 0.023 215.469 0.000 4.879 4.969
===================================================================================
Ljung-Box (L1) (Q): 14.55 Jarque-Bera (JB): 2468024.38
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 274.15
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.06158, saving model to LSTM2.h5
81/81 - 6s - loss: 0.1058 - accuracy: 0.0000e+00 - val_loss: 0.0616 - val_accuracy: 0.0037 - lr: 0.0010 - 6s/epoch - 71ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.06158 to 0.04372, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0206 - accuracy: 0.0000e+00 - val_loss: 0.0437 - val_accuracy: 0.0037 - lr: 0.0010 - 795ms/epoch - 10ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.04372 to 0.01454, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0351 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 0.0010 - 728ms/epoch - 9ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.01454 to 0.00900, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0366 - accuracy: 0.0000e+00 - val_loss: 0.0090 - val_accuracy: 0.0037 - lr: 0.0010 - 785ms/epoch - 10ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.00900 to 0.00659, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0315 - accuracy: 0.0000e+00 - val_loss: 0.0066 - val_accuracy: 0.0037 - lr: 0.0010 - 741ms/epoch - 9ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.00659 to 0.00451, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0242 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 0.0010 - 715ms/epoch - 9ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.00451 to 0.00359, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0152 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 0.0010 - 735ms/epoch - 9ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.00359 to 0.00329, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0104 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 0.0010 - 781ms/epoch - 10ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00329
81/81 - 1s - loss: 0.0085 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 0.0010 - 684ms/epoch - 8ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00329
81/81 - 1s - loss: 0.0086 - accuracy: 0.0000e+00 - val_loss: 0.0038 - val_accuracy: 0.0037 - lr: 0.0010 - 705ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.00329
81/81 - 1s - loss: 0.0097 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 0.0010 - 702ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00329
81/81 - 1s - loss: 0.0117 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 0.0010 - 708ms/epoch - 9ms/step
Epoch 13/500
Epoch 00013: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00013: val_loss did not improve from 0.00329
81/81 - 1s - loss: 0.0130 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 0.0010 - 697ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00329
81/81 - 1s - loss: 0.0180 - accuracy: 0.0000e+00 - val_loss: 0.0052 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 741ms/epoch - 9ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00329
81/81 - 1s - loss: 0.0043 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 702ms/epoch - 9ms/step
Epoch 16/500
Epoch 00016: val_loss improved from 0.00329 to 0.00301, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0030 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 757ms/epoch - 9ms/step
Epoch 17/500
Epoch 00017: val_loss improved from 0.00301 to 0.00264, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0021 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 707ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss improved from 0.00264 to 0.00248, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0016 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 789ms/epoch - 10ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00248
81/81 - 1s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 713ms/epoch - 9ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00248
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 682ms/epoch - 8ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00248
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 744ms/epoch - 9ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00248
81/81 - 1s - loss: 9.3281e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 711ms/epoch - 9ms/step
Epoch 23/500
Epoch 00023: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00023: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.9032e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 693ms/epoch - 9ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.4976e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 748ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.4727e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 683ms/epoch - 8ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.4534e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 686ms/epoch - 8ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.4359e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 693ms/epoch - 9ms/step
Epoch 28/500
Epoch 00028: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00028: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.4192e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 690ms/epoch - 9ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.4028e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 725ms/epoch - 9ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.3865e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 704ms/epoch - 9ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.3703e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.3541e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 686ms/epoch - 8ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.3379e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 679ms/epoch - 8ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.3216e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 692ms/epoch - 9ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.3053e-04 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 753ms/epoch - 9ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.2890e-04 - accuracy: 0.0000e+00 - val_loss: 0.0048 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 711ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.2727e-04 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.2564e-04 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 770ms/epoch - 10ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.2401e-04 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.2238e-04 - accuracy: 0.0000e+00 - val_loss: 0.0051 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.2075e-04 - accuracy: 0.0000e+00 - val_loss: 0.0052 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 696ms/epoch - 9ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.1911e-04 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 710ms/epoch - 9ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.1749e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 712ms/epoch - 9ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.1586e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 702ms/epoch - 9ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.1423e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 704ms/epoch - 9ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.1260e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 698ms/epoch - 9ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.1097e-04 - accuracy: 0.0000e+00 - val_loss: 0.0057 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 701ms/epoch - 9ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.0933e-04 - accuracy: 0.0000e+00 - val_loss: 0.0058 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 744ms/epoch - 9ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.0770e-04 - accuracy: 0.0000e+00 - val_loss: 0.0059 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.0606e-04 - accuracy: 0.0000e+00 - val_loss: 0.0060 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 724ms/epoch - 9ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.0442e-04 - accuracy: 0.0000e+00 - val_loss: 0.0062 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 681ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.0277e-04 - accuracy: 0.0000e+00 - val_loss: 0.0063 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 682ms/epoch - 8ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00248
81/81 - 1s - loss: 8.0112e-04 - accuracy: 0.0000e+00 - val_loss: 0.0064 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 703ms/epoch - 9ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.9946e-04 - accuracy: 0.0000e+00 - val_loss: 0.0065 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 709ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.9780e-04 - accuracy: 0.0000e+00 - val_loss: 0.0066 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 702ms/epoch - 9ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.9612e-04 - accuracy: 0.0000e+00 - val_loss: 0.0067 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.9444e-04 - accuracy: 0.0000e+00 - val_loss: 0.0068 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 687ms/epoch - 8ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.9274e-04 - accuracy: 0.0000e+00 - val_loss: 0.0069 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.9104e-04 - accuracy: 0.0000e+00 - val_loss: 0.0070 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 717ms/epoch - 9ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.8932e-04 - accuracy: 0.0000e+00 - val_loss: 0.0071 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 708ms/epoch - 9ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.8759e-04 - accuracy: 0.0000e+00 - val_loss: 0.0072 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 703ms/epoch - 9ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.8584e-04 - accuracy: 0.0000e+00 - val_loss: 0.0073 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 713ms/epoch - 9ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.8408e-04 - accuracy: 0.0000e+00 - val_loss: 0.0074 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 710ms/epoch - 9ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.8231e-04 - accuracy: 0.0000e+00 - val_loss: 0.0075 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 686ms/epoch - 8ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.8051e-04 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 699ms/epoch - 9ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.7870e-04 - accuracy: 0.0000e+00 - val_loss: 0.0077 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 694ms/epoch - 9ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.7688e-04 - accuracy: 0.0000e+00 - val_loss: 0.0078 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.00248
81/81 - 1s - loss: 7.7503e-04 - accuracy: 0.0000e+00 - val_loss: 0.0079 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 706ms/epoch - 9ms/step
Epoch 00068: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 136.8787627980392
RMSE: 11.699519767838302
MAPE: 9.782684450137419
EMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 44.78% Accuracy
MSE: 325.2876788257106
RMSE: 18.035733387520192
MAPE: 15.58326491461623
WMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.28% Accuracy
MSE: 142.26654445774975
RMSE: 11.927554001460221
MAPE: 9.787176512242624
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 171.26734938505615
RMSE: 13.086915197442679
MAPE: 11.821213958102536
KAMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 52.24625253706662
RMSE: 7.228156925321048
MAPE: 5.787764162926424
MIDPOINT
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 51.03620830548878
RMSE: 7.143963067197981
MAPE: 5.7689786092745114
T3
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 48.13% Accuracy
MSE: 131.82692716984482
RMSE: 11.481590794391028
MAPE: 9.148826908925223
TEMA
TEMA([input_arrays], [timeperiod=30])
Triple Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
9
Working on TEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.66 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4352.703, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3889.412, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.38 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3689.930, Time=0.08 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3574.245, Time=0.11 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.50 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.04 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3576.245, Time=0.28 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.165 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1783.123
Date: Sun, 12 Dec 2021 AIC 3574.245
Time: 12:34:50 BIC 3593.008
Sample: 0 HQIC 3581.451
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1480 0.004 -302.430 0.000 -1.155 -1.141
ar.L2 -0.8300 0.008 -99.682 0.000 -0.846 -0.814
ar.L3 -0.3687 0.007 -50.527 0.000 -0.383 -0.354
sigma2 4.9055 0.028 175.970 0.000 4.851 4.960
===================================================================================
Ljung-Box (L1) (Q): 11.61 Jarque-Bera (JB): 1261976.58
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.16 Skew: 2.52
Prob(H) (two-sided): 0.00 Kurtosis: 196.90
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.09338, saving model to LSTM2.h5
81/81 - 6s - loss: 0.0863 - accuracy: 0.0000e+00 - val_loss: 0.0934 - val_accuracy: 0.0037 - lr: 0.0010 - 6s/epoch - 77ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.09338 to 0.01889, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0719 - accuracy: 0.0000e+00 - val_loss: 0.0189 - val_accuracy: 0.0037 - lr: 0.0010 - 715ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.01889 to 0.00405, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0139 - accuracy: 0.0000e+00 - val_loss: 0.0040 - val_accuracy: 0.0037 - lr: 0.0010 - 728ms/epoch - 9ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.00405
81/81 - 1s - loss: 0.0188 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 0.0010 - 745ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.00405
81/81 - 1s - loss: 0.0208 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 0.0010 - 694ms/epoch - 9ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.00405
81/81 - 1s - loss: 0.0166 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 0.0010 - 676ms/epoch - 8ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.00405
81/81 - 1s - loss: 0.0150 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 0.0010 - 728ms/epoch - 9ms/step
Epoch 8/500
Epoch 00008: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00008: val_loss improved from 0.00405 to 0.00395, saving model to LSTM2.h5
81/81 - 1s - loss: 0.0148 - accuracy: 0.0000e+00 - val_loss: 0.0040 - val_accuracy: 0.0037 - lr: 0.0010 - 747ms/epoch - 9ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0191 - accuracy: 0.0000e+00 - val_loss: 0.0226 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 674ms/epoch - 8ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0039 - accuracy: 0.0000e+00 - val_loss: 0.0188 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 709ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0028 - accuracy: 0.0000e+00 - val_loss: 0.0158 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 694ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0020 - accuracy: 0.0000e+00 - val_loss: 0.0131 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 680ms/epoch - 8ms/step
Epoch 13/500
Epoch 00013: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00013: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0015 - accuracy: 0.0000e+00 - val_loss: 0.0107 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 693ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0107 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 681ms/epoch - 8ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0107 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 709ms/epoch - 9ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0106 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 676ms/epoch - 8ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0105 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 732ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00018: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0103 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 716ms/epoch - 9ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0101 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 694ms/epoch - 9ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0099 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 712ms/epoch - 9ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0096 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 699ms/epoch - 9ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0094 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 731ms/epoch - 9ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0091 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 711ms/epoch - 9ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0089 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 699ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0086 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 698ms/epoch - 9ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00395
81/81 - 1s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0084 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 693ms/epoch - 9ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.9689e-04 - accuracy: 0.0000e+00 - val_loss: 0.0081 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 706ms/epoch - 9ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.8528e-04 - accuracy: 0.0000e+00 - val_loss: 0.0079 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 748ms/epoch - 9ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.7432e-04 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 692ms/epoch - 9ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.6402e-04 - accuracy: 0.0000e+00 - val_loss: 0.0074 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 702ms/epoch - 9ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.5435e-04 - accuracy: 0.0000e+00 - val_loss: 0.0071 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 754ms/epoch - 9ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.4532e-04 - accuracy: 0.0000e+00 - val_loss: 0.0069 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 753ms/epoch - 9ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.3690e-04 - accuracy: 0.0000e+00 - val_loss: 0.0066 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 725ms/epoch - 9ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.2908e-04 - accuracy: 0.0000e+00 - val_loss: 0.0064 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 712ms/epoch - 9ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.2183e-04 - accuracy: 0.0000e+00 - val_loss: 0.0061 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 718ms/epoch - 9ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.1511e-04 - accuracy: 0.0000e+00 - val_loss: 0.0059 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 706ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.0890e-04 - accuracy: 0.0000e+00 - val_loss: 0.0057 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 698ms/epoch - 9ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00395
81/81 - 1s - loss: 9.0317e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 706ms/epoch - 9ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00395
81/81 - 1s - loss: 8.9787e-04 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 684ms/epoch - 8ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00395
81/81 - 1s - loss: 8.9298e-04 - accuracy: 0.0000e+00 - val_loss: 0.0051 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 700ms/epoch - 9ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00395
81/81 - 1s - loss: 8.8845e-04 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 706ms/epoch - 9ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00395
81/81 - 1s - loss: 8.8424e-04 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 700ms/epoch - 9ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00395
81/81 - 1s - loss: 8.8034e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 707ms/epoch - 9ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00395
81/81 - 1s - loss: 8.7669e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 706ms/epoch - 9ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00395
81/81 - 1s - loss: 8.7328e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 729ms/epoch - 9ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00395
81/81 - 1s - loss: 8.7006e-04 - accuracy: 0.0000e+00 - val_loss: 0.0040 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 703ms/epoch - 9ms/step
Epoch 47/500
Epoch 00047: val_loss improved from 0.00395 to 0.00387, saving model to LSTM2.h5
81/81 - 1s - loss: 8.6701e-04 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 909ms/epoch - 11ms/step
Epoch 48/500
Epoch 00048: val_loss improved from 0.00387 to 0.00372, saving model to LSTM2.h5
81/81 - 1s - loss: 8.6411e-04 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 761ms/epoch - 9ms/step
Epoch 49/500
Epoch 00049: val_loss improved from 0.00372 to 0.00359, saving model to LSTM2.h5
81/81 - 1s - loss: 8.6133e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 728ms/epoch - 9ms/step
Epoch 50/500
Epoch 00050: val_loss improved from 0.00359 to 0.00347, saving model to LSTM2.h5
81/81 - 1s - loss: 8.5865e-04 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 716ms/epoch - 9ms/step
Epoch 51/500
Epoch 00051: val_loss improved from 0.00347 to 0.00335, saving model to LSTM2.h5
81/81 - 1s - loss: 8.5605e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 728ms/epoch - 9ms/step
Epoch 52/500
Epoch 00052: val_loss improved from 0.00335 to 0.00324, saving model to LSTM2.h5
81/81 - 1s - loss: 8.5352e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 734ms/epoch - 9ms/step
Epoch 53/500
Epoch 00053: val_loss improved from 0.00324 to 0.00314, saving model to LSTM2.h5
81/81 - 1s - loss: 8.5104e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 776ms/epoch - 10ms/step
Epoch 54/500
Epoch 00054: val_loss improved from 0.00314 to 0.00304, saving model to LSTM2.h5
81/81 - 1s - loss: 8.4859e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 722ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss improved from 0.00304 to 0.00296, saving model to LSTM2.h5
81/81 - 1s - loss: 8.4617e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 731ms/epoch - 9ms/step
Epoch 56/500
Epoch 00056: val_loss improved from 0.00296 to 0.00288, saving model to LSTM2.h5
81/81 - 1s - loss: 8.4377e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 746ms/epoch - 9ms/step
Epoch 57/500
Epoch 00057: val_loss improved from 0.00288 to 0.00280, saving model to LSTM2.h5
81/81 - 1s - loss: 8.4138e-04 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 748ms/epoch - 9ms/step
Epoch 58/500
Epoch 00058: val_loss improved from 0.00280 to 0.00274, saving model to LSTM2.h5
81/81 - 1s - loss: 8.3899e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 727ms/epoch - 9ms/step
Epoch 59/500
Epoch 00059: val_loss improved from 0.00274 to 0.00268, saving model to LSTM2.h5
81/81 - 1s - loss: 8.3660e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 736ms/epoch - 9ms/step
Epoch 60/500
Epoch 00060: val_loss improved from 0.00268 to 0.00262, saving model to LSTM2.h5
81/81 - 1s - loss: 8.3421e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 769ms/epoch - 9ms/step
Epoch 61/500
Epoch 00061: val_loss improved from 0.00262 to 0.00258, saving model to LSTM2.h5
81/81 - 1s - loss: 8.3181e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 755ms/epoch - 9ms/step
Epoch 62/500
Epoch 00062: val_loss improved from 0.00258 to 0.00254, saving model to LSTM2.h5
81/81 - 1s - loss: 8.2940e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 736ms/epoch - 9ms/step
Epoch 63/500
Epoch 00063: val_loss improved from 0.00254 to 0.00250, saving model to LSTM2.h5
81/81 - 1s - loss: 8.2697e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 729ms/epoch - 9ms/step
Epoch 64/500
Epoch 00064: val_loss improved from 0.00250 to 0.00248, saving model to LSTM2.h5
81/81 - 1s - loss: 8.2454e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 783ms/epoch - 10ms/step
Epoch 65/500
Epoch 00065: val_loss improved from 0.00248 to 0.00246, saving model to LSTM2.h5
81/81 - 1s - loss: 8.2209e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 750ms/epoch - 9ms/step
Epoch 66/500
Epoch 00066: val_loss improved from 0.00246 to 0.00244, saving model to LSTM2.h5
81/81 - 1s - loss: 8.1963e-04 - accuracy: 0.0000e+00 - val_loss: 0.0024 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 707ms/epoch - 9ms/step
Epoch 67/500
Epoch 00067: val_loss improved from 0.00244 to 0.00243, saving model to LSTM2.h5
81/81 - 1s - loss: 8.1715e-04 - accuracy: 0.0000e+00 - val_loss: 0.0024 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 739ms/epoch - 9ms/step
Epoch 68/500
Epoch 00068: val_loss improved from 0.00243 to 0.00243, saving model to LSTM2.h5
81/81 - 1s - loss: 8.1466e-04 - accuracy: 0.0000e+00 - val_loss: 0.0024 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 735ms/epoch - 9ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.00243
81/81 - 1s - loss: 8.1216e-04 - accuracy: 0.0000e+00 - val_loss: 0.0024 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 693ms/epoch - 9ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.00243
81/81 - 1s - loss: 8.0964e-04 - accuracy: 0.0000e+00 - val_loss: 0.0024 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 707ms/epoch - 9ms/step
Epoch 71/500
Epoch 00071: val_loss did not improve from 0.00243
81/81 - 1s - loss: 8.0712e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 711ms/epoch - 9ms/step
Epoch 72/500
Epoch 00072: val_loss did not improve from 0.00243
81/81 - 1s - loss: 8.0459e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 684ms/epoch - 8ms/step
Epoch 73/500
Epoch 00073: val_loss did not improve from 0.00243
81/81 - 1s - loss: 8.0204e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 74/500
Epoch 00074: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.9949e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 704ms/epoch - 9ms/step
Epoch 75/500
Epoch 00075: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.9693e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 718ms/epoch - 9ms/step
Epoch 76/500
Epoch 00076: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.9437e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 679ms/epoch - 8ms/step
Epoch 77/500
Epoch 00077: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.9180e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 691ms/epoch - 9ms/step
Epoch 78/500
Epoch 00078: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.8923e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 721ms/epoch - 9ms/step
Epoch 79/500
Epoch 00079: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.8666e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 80/500
Epoch 00080: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.8408e-04 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 700ms/epoch - 9ms/step
Epoch 81/500
Epoch 00081: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.8150e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 703ms/epoch - 9ms/step
Epoch 82/500
Epoch 00082: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.7893e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 712ms/epoch - 9ms/step
Epoch 83/500
Epoch 00083: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.7635e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 84/500
Epoch 00084: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.7378e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 696ms/epoch - 9ms/step
Epoch 85/500
Epoch 00085: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.7121e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 701ms/epoch - 9ms/step
Epoch 86/500
Epoch 00086: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.6864e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 743ms/epoch - 9ms/step
Epoch 87/500
Epoch 00087: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.6608e-04 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 716ms/epoch - 9ms/step
Epoch 88/500
Epoch 00088: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.6352e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 702ms/epoch - 9ms/step
Epoch 89/500
Epoch 00089: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.6097e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 694ms/epoch - 9ms/step
Epoch 90/500
Epoch 00090: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.5843e-04 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 712ms/epoch - 9ms/step
Epoch 91/500
Epoch 00091: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.5589e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 710ms/epoch - 9ms/step
Epoch 92/500
Epoch 00092: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.5336e-04 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 712ms/epoch - 9ms/step
Epoch 93/500
Epoch 00093: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.5084e-04 - accuracy: 0.0000e+00 - val_loss: 0.0038 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 706ms/epoch - 9ms/step
Epoch 94/500
Epoch 00094: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.4834e-04 - accuracy: 0.0000e+00 - val_loss: 0.0038 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 717ms/epoch - 9ms/step
Epoch 95/500
Epoch 00095: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.4584e-04 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 689ms/epoch - 9ms/step
Epoch 96/500
Epoch 00096: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.4335e-04 - accuracy: 0.0000e+00 - val_loss: 0.0040 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 699ms/epoch - 9ms/step
Epoch 97/500
Epoch 00097: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.4088e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 688ms/epoch - 8ms/step
Epoch 98/500
Epoch 00098: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.3842e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 698ms/epoch - 9ms/step
Epoch 99/500
Epoch 00099: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.3597e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 686ms/epoch - 8ms/step
Epoch 100/500
Epoch 00100: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.3354e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 101/500
Epoch 00101: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.3112e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 711ms/epoch - 9ms/step
Epoch 102/500
Epoch 00102: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.2871e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 714ms/epoch - 9ms/step
Epoch 103/500
Epoch 00103: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.2632e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 718ms/epoch - 9ms/step
Epoch 104/500
Epoch 00104: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.2395e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 744ms/epoch - 9ms/step
Epoch 105/500
Epoch 00105: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.2160e-04 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 741ms/epoch - 9ms/step
Epoch 106/500
Epoch 00106: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.1926e-04 - accuracy: 0.0000e+00 - val_loss: 0.0048 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 692ms/epoch - 9ms/step
Epoch 107/500
Epoch 00107: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.1694e-04 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 766ms/epoch - 9ms/step
Epoch 108/500
Epoch 00108: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.1464e-04 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 728ms/epoch - 9ms/step
Epoch 109/500
Epoch 00109: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.1236e-04 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 701ms/epoch - 9ms/step
Epoch 110/500
Epoch 00110: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.1010e-04 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 741ms/epoch - 9ms/step
Epoch 111/500
Epoch 00111: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.0786e-04 - accuracy: 0.0000e+00 - val_loss: 0.0051 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 725ms/epoch - 9ms/step
Epoch 112/500
Epoch 00112: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.0564e-04 - accuracy: 0.0000e+00 - val_loss: 0.0052 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 704ms/epoch - 9ms/step
Epoch 113/500
Epoch 00113: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.0344e-04 - accuracy: 0.0000e+00 - val_loss: 0.0052 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 695ms/epoch - 9ms/step
Epoch 114/500
Epoch 00114: val_loss did not improve from 0.00243
81/81 - 1s - loss: 7.0126e-04 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 705ms/epoch - 9ms/step
Epoch 115/500
Epoch 00115: val_loss did not improve from 0.00243
81/81 - 1s - loss: 6.9910e-04 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 734ms/epoch - 9ms/step
Epoch 116/500
Epoch 00116: val_loss did not improve from 0.00243
81/81 - 1s - loss: 6.9696e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 719ms/epoch - 9ms/step
Epoch 117/500
Epoch 00117: val_loss did not improve from 0.00243
81/81 - 1s - loss: 6.9484e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 742ms/epoch - 9ms/step
Epoch 118/500
Epoch 00118: val_loss did not improve from 0.00243
81/81 - 1s - loss: 6.9275e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 749ms/epoch - 9ms/step
Epoch 00118: early stopping
SMA Prediction vs Close: 52.61% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 136.8787627980392 RMSE: 11.699519767838302 MAPE: 9.782684450137419 EMA Prediction vs Close: 52.24% Accuracy Prediction vs Prediction: 44.78% Accuracy MSE: 325.2876788257106 RMSE: 18.035733387520192 MAPE: 15.58326491461623 WMA Prediction vs Close: 54.85% Accuracy Prediction vs Prediction: 43.28% Accuracy MSE: 142.26654445774975 RMSE: 11.927554001460221 MAPE: 9.787176512242624 DEMA Prediction vs Close: 52.24% Accuracy Prediction vs Prediction: 45.52% Accuracy MSE: 171.26734938505615 RMSE: 13.086915197442679 MAPE: 11.821213958102536 KAMA Prediction vs Close: 52.24% Accuracy Prediction vs Prediction: 47.01% Accuracy MSE: 52.24625253706662 RMSE: 7.228156925321048 MAPE: 5.787764162926424 MIDPOINT Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 45.52% Accuracy MSE: 51.03620830548878 RMSE: 7.143963067197981 MAPE: 5.7689786092745114 T3 Prediction vs Close: 54.48% Accuracy Prediction vs Prediction: 48.13% Accuracy MSE: 131.82692716984482 RMSE: 11.481590794391028 MAPE: 9.148826908925223 TEMA Prediction vs Close: 50.37% Accuracy Prediction vs Prediction: 48.51% Accuracy MSE: 103.01748696218638 RMSE: 10.149753049320283 MAPE: 9.012271728014667 Runtime: mins: 20.660532839866665
Architecture Used¶
In [ ]:
from google.colab import files
import cv2
uploaded = files.upload()
Saving Experiment2.png to Experiment2 (1).png
In [ ]:
img = cv2.imread('Experiment2.png')
plt.figure(figsize=(20,10))
plt.axis("off")
plt.title('LSTM Architecture '+imgfile,fontsize=18)
plt.imshow(img)
Out[ ]:
<matplotlib.image.AxesImage at 0x7fdfb2c6f290>
Model Plots¶
In [73]:
with open('simulation2_data.json') as json_file:
simulation2 = json.load(json_file)
fileimg = 'Experiment2'
In [74]:
for i in range(len(list(simulation2.keys()))):
SIM = list(simulation2.keys())[i]
plot_train(simulation2,SIM)
plot_test(simulation2,SIM)
----- Train RMSE for SMA ----- 8.826942967936205 ----- Train_MSE_LSTM for SMA ----- 77.91492215919841 ----- Train MAE LSTM for SMA ----- 7.736753512883909
----- Test RMSE for SMA----- 11.699519767838302 ----- Test_MSE_LSTM for SMA----- 136.8787627980392 ----- Test_MAE_LSTM for SMA----- 9.782684450137419
----- Train RMSE for EMA ----- 9.86405135909782 ----- Train_MSE_LSTM for EMA ----- 97.29950921491954 ----- Train MAE LSTM for EMA ----- 8.73726280728392
----- Test RMSE for EMA----- 18.035733387520192 ----- Test_MSE_LSTM for EMA----- 325.2876788257106 ----- Test_MAE_LSTM for EMA----- 15.58326491461623
----- Train RMSE for WMA ----- 10.453164264775387 ----- Train_MSE_LSTM for WMA ----- 109.26864314637714 ----- Train MAE LSTM for WMA ----- 9.361118262029704
----- Test RMSE for WMA----- 11.927554001460221 ----- Test_MSE_LSTM for WMA----- 142.26654445774975 ----- Test_MAE_LSTM for WMA----- 9.787176512242624
----- Train RMSE for DEMA ----- 12.156210755746763 ----- Train_MSE_LSTM for DEMA ----- 147.77345993813327 ----- Train MAE LSTM for DEMA ----- 10.956398825846396
----- Test RMSE for DEMA----- 13.086915197442679 ----- Test_MSE_LSTM for DEMA----- 171.26734938505615 ----- Test_MAE_LSTM for DEMA----- 11.821213958102536
----- Train RMSE for KAMA ----- 10.497854644863434 ----- Train_MSE_LSTM for KAMA ----- 110.20495214468077 ----- Train MAE LSTM for KAMA ----- 9.473006183673434
----- Test RMSE for KAMA----- 7.228156925321048 ----- Test_MSE_LSTM for KAMA----- 52.24625253706662 ----- Test_MAE_LSTM for KAMA----- 5.787764162926424
----- Train RMSE for MIDPOINT ----- 9.46823580889301 ----- Train_MSE_LSTM for MIDPOINT ----- 89.64748933280386 ----- Train MAE LSTM for MIDPOINT ----- 8.431252318872415
----- Test RMSE for MIDPOINT----- 7.143963067197981 ----- Test_MSE_LSTM for MIDPOINT----- 51.03620830548878 ----- Test_MAE_LSTM for MIDPOINT----- 5.7689786092745114
----- Train RMSE for T3 ----- 12.040312964311324 ----- Train_MSE_LSTM for T3 ----- 144.96913627856335 ----- Train MAE LSTM for T3 ----- 10.855980292569482
----- Test RMSE for T3----- 11.481590794391028 ----- Test_MSE_LSTM for T3----- 131.82692716984482 ----- Test_MAE_LSTM for T3----- 9.148826908925223
----- Train RMSE for TEMA ----- 7.410384228353796 ----- Train_MSE_LSTM for TEMA ----- 54.91379441183469 ----- Train MAE LSTM for TEMA ----- 5.102082156767552
----- Test RMSE for TEMA----- 10.149753049320283 ----- Test_MSE_LSTM for TEMA----- 103.01748696218638 ----- Test_MAE_LSTM for TEMA----- 9.012271728014667
Univariate Arima Multistep MutiVariate LSTM Hybrid Model Experiment 3¶
In [ ]:
def get_lstm(data,original_data, train_len, test_len,img_file,ma ,lstm_len=3):
# prepare train and test data
X_value = pd.DataFrame(data.iloc[:, :])
y_value = pd.DataFrame(data.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
# Get data and check shape
X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X)
y_train, y_test, = split_train_test(y)
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
det = 20
input_dim = X_train.shape[1]#3
feature_size = X_train.shape[2]#24
output_dim = y_train.shape[1]#1
# # Option 1
# # Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(256, activation='relu', kernel_initializer='he_normal', input_shape=(input_dim, feature_size)))
# model.add(Dense(units=64,activation='relu'))
# model.add(Dropout(0.5))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(learning_rate = 0.001), loss='mse')
# ## Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# # option 2
# model = Sequential()
# model.add(Bidirectional(LSTM(units= 128), input_shape=(input_dim, feature_size)))
# model.add(Dense(64))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(lr = 0.001), loss='mean_squared_error', metrics=['accuracy'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Option 3
# define custom activation
#
class Double_Tanh(Activation):
def __init__(self, activation, **kwargs):
super(Double_Tanh, self).__init__(activation, **kwargs)
self.__name__ = 'double_tanh'
def double_tanh(x):
return (K.tanh(x) * 2)
get_custom_objects().update({'double_tanh':Double_Tanh(double_tanh)})
# Model Generation
model = Sequential()
#check https://machinelearningmastery.com/use-weight-regularization-lstm-networks-time-series-forecasting/
model.add(LSTM(25, input_shape=(input_dim, feature_size), dropout=0.2, kernel_regularizer=l1_l2(0.00,0.00), bias_regularizer=l1_l2(0.00,0.00)))
model.add(Dense(1))
model.add(Activation(double_tanh))
model.compile(loss='mean_squared_error', optimizer='adam', metrics=['mse', 'mae'])
# Common code
callbacks = [
EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
ModelCheckpoint('LSTM3.h5', verbose=1, save_best_only=True, save_weights_only=True)]
fname1 = img_file+'.png'
tensorflow.keras.utils.plot_model(
model, to_file=fname1, show_shapes=True, show_dtype=False,
show_layer_names=True, expand_nested=False, dpi=96,
layer_range=None, show_layer_activations=False
)
history = model.fit(X_train, y_train, epochs=500, batch_size=int( optimized_period[ma]), verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# plot loss
fname2 = img_file+'-'+ma
plt.title(img_file+'-'+ma+' Loss')
plt.xlabel("Epochs")
plt.ylabel("Loss")
pyplot.plot(history.history['loss'], label='train')
pyplot.plot(history.history['val_loss'], label='validation')
pyplot.legend()
pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# Option 4
# Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(units=lstm_len, return_sequences=True, input_shape=(x_train.shape[1], 1)))
# model.add(LSTM(units=int(lstm_len/2)))
# model.add(Dense(1, activation='sigmoid'))
# model.compile(loss='mean_squared_error', optimizer='adam')
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Generate predictions
predictiontr = model.predict(X_train, verbose=0)
predictiontr = y_scaler.inverse_transform(predictiontr).tolist()
outputtr = []
for i in range(len(predictiontr)):
outputtr.extend(predictiontr[i])
predictiontr = outputtr
# Generate error data
## replace with yc , xtest generated by new multistep method
mse_tr = mean_squared_error(y_train, predictiontr)
rmse_tr = mse_tr ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictiontr))
mae_tr = mean_absolute_error(y_train, pd.Series(predictiontr))
# Original_tr = pd.Series(yc_train)
Original_tr = y_scaler.inverse_transform(y_train).flatten().tolist()
predictionte = model.predict(X_test, verbose=0)
predictionte = (y_scaler.inverse_transform(predictionte)-det).tolist()
outputte = []
for i in range(len(predictionte)):
outputte.extend(predictionte[i])
predictionte = outputte
# Generate error data
mse_te = mean_squared_error(y_test, predictionte)
rmse_te = mse_te ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictionte))
mae_te = mean_absolute_error(y_test, pd.Series(predictionte))
# Original_te = pd.Series(yc_test)
Original_te = y_scaler.inverse_transform(y_test).flatten().tolist()
return Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,Original_te,predictionte, mse_te,rmse_te,mae_te
In [ ]:
if __name__ == '__main__':
start_time = timeit.default_timer()
simulation3 = {}
imgfile = 'Experiment3'
for ma in optimized_period:
print(ma)
print(functions[ma])
print ( int( optimized_period[ma]))
# if ma == 'SMA':
low_vol = df.apply(lambda c: functions[ma](c, timeperiod = int( optimized_period[ma])))
low_vol = low_vol.fillna(0)
low_vol_data = df['close']
high_vol = pd.DataFrame()
df2 = df.copy()
for i in df2.columns:
if i in low_vol.columns:
high_vol[i] = df2[i].subtract(low_vol[i], fill_value=0)
high_vol_data = df['close']
## *****************************************************
# Generate ARIMA and LSTM predictions
print('\nWorking on ' + ma + ' predictions')
try:
print('parameters used : ', train_len, test_len)
low_vol_Original, low_vol_prediction, low_vol_mse, low_vol_rmse,low_vol_mae = get_arima(low_vol,low_vol_data, train_len, test_len)
except:
print('ARIMA error, skipping to next MA type')
continue
Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,high_vol_Original, high_vol_prediction, high_vol_mse, high_vol_rmse,high_vol_mae, = get_lstm(high_vol,high_vol_data, train_len, test_len,imgfile,ma)
final_prediction_tr = df['close'].head(train_len).values + pd.Series(predictiontr) # ignoring first 3 steps
mse_ftr = mean_squared_error(df['close'].head(train_len).values,final_prediction_tr.values)
rmse_ftr = mse_ftr ** 0.5
mape_ftr = mean_absolute_percentage_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
mae_ftr = mean_absolute_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
final_prediction = pd.Series(low_vol_prediction[3:]) + pd.Series(high_vol_prediction)
mse = mean_squared_error(df['close'].tail(test_len).values,final_prediction.values)
rmse = mse ** 0.5
mape = mean_absolute_percentage_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
mae = mean_absolute_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
# Generate prediction accuracy
actual = df['close'].tail(test_len).values
result_1 = []
result_2 = []
for i in range(1, len(final_prediction)):
# Compare prediction to previous close price
if final_prediction[i] > actual[i-1] and actual[i] > actual[i-1]:
result_1.append(1)
elif final_prediction[i] < actual[i-1] and actual[i] < actual[i-1]:
result_1.append(1)
else:
result_1.append(0)
# Compare prediction to previous prediction
if final_prediction[i] > final_prediction[i-1] and actual[i] > actual[i-1]:
result_2.append(1)
elif final_prediction[i] < final_prediction[i-1] and actual[i] < actual[i-1]:
result_2.append(1)
else:
result_2.append(0)
accuracy_1 = np.mean(result_1)
accuracy_2 = np.mean(result_2)
simulation3[ma] = {'low_vol': {'original':list(low_vol_Original), 'prediction': list(low_vol_prediction) , 'mse': low_vol_mse,
'rmse': low_vol_rmse, 'mae' : low_vol_mae},
'high_vol': {'original':list(high_vol_Original),'prediction': list(high_vol_prediction), 'mse': high_vol_mse,
'rmse': high_vol_rmse, 'mae' : high_vol_mae},
'final_tr': {'original':df['close'].head(train_len).tolist(),'prediction': final_prediction_tr.values.tolist(), 'mse': mse_ftr,
'rmse': rmse_ftr, 'mae' : mae_ftr},
'final': {'original': df['close'].tail(test_len).tolist(), 'prediction': final_prediction.values.tolist(), 'mse': mse,
'rmse': rmse, 'mae': mae },
'accuracy': {'prediction vs close': accuracy_1, 'prediction vs prediction': accuracy_2}}
# save simulation data here as checkpoint
with open('simulation3_data.json', 'w') as fp:
json.dump(simulation3, fp)
for ma in simulation3.keys():
print('\n' + ma)
print('Prediction vs Close:\t\t' + str(round(100*simulation3[ma]['accuracy']['prediction vs close'], 2))
+ '% Accuracy')
print('Prediction vs Prediction:\t' + str(round(100*simulation3[ma]['accuracy']['prediction vs prediction'], 2))
+ '% Accuracy')
print('MSE:\t', simulation3[ma]['final']['mse'],
'\nRMSE:\t', simulation3[ma]['final']['rmse'],
'\nMAPE:\t', simulation3[ma]['final']['mae'])#,
# '\nMAPE:\t', simulation[ma]['final']['mape'])
# else:
# break
elapsed = timeit.default_timer() - start_time
print('Runtime: mins:',elapsed/60)
SMA
SMA([input_arrays], [timeperiod=30])
Simple Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
17
Working on SMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.68 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4157.020, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3687.148, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.28 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3458.651, Time=0.09 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3322.133, Time=0.12 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=0.97 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.05 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3324.133, Time=0.27 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 3.581 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1657.067
Date: Sun, 12 Dec 2021 AIC 3322.133
Time: 12:40:44 BIC 3340.897
Sample: 0 HQIC 3329.339
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1966 0.003 -387.226 0.000 -1.203 -1.191
ar.L2 -0.8952 0.006 -138.692 0.000 -0.908 -0.883
ar.L3 -0.3968 0.006 -68.284 0.000 -0.408 -0.385
sigma2 3.5858 0.017 214.535 0.000 3.553 3.619
===================================================================================
Ljung-Box (L1) (Q): 14.47 Jarque-Bera (JB): 2428881.42
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 271.99
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.36232, saving model to LSTM3.h5
48/48 - 3s - loss: 0.2069 - mse: 0.2069 - mae: 0.3363 - val_loss: 0.3623 - val_mse: 0.3623 - val_mae: 0.5924 - lr: 0.0010 - 3s/epoch - 65ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.36232 to 0.17260, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0285 - mse: 0.0285 - mae: 0.1319 - val_loss: 0.1726 - val_mse: 0.1726 - val_mae: 0.4047 - lr: 0.0010 - 353ms/epoch - 7ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.17260 to 0.08198, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0180 - mse: 0.0180 - mae: 0.1071 - val_loss: 0.0820 - val_mse: 0.0820 - val_mae: 0.2728 - lr: 0.0010 - 367ms/epoch - 8ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.08198 to 0.05134, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.1013 - val_loss: 0.0513 - val_mse: 0.0513 - val_mae: 0.2104 - lr: 0.0010 - 333ms/epoch - 7ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.05134 to 0.03382, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0141 - mse: 0.0141 - mae: 0.0960 - val_loss: 0.0338 - val_mse: 0.0338 - val_mae: 0.1655 - lr: 0.0010 - 341ms/epoch - 7ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.03382 to 0.02675, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0142 - mse: 0.0142 - mae: 0.0954 - val_loss: 0.0267 - val_mse: 0.0267 - val_mae: 0.1439 - lr: 0.0010 - 343ms/epoch - 7ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.02675
48/48 - 0s - loss: 0.0124 - mse: 0.0124 - mae: 0.0894 - val_loss: 0.0276 - val_mse: 0.0276 - val_mae: 0.1466 - lr: 0.0010 - 341ms/epoch - 7ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.02675 to 0.02619, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0143 - mse: 0.0143 - mae: 0.0972 - val_loss: 0.0262 - val_mse: 0.0262 - val_mae: 0.1422 - lr: 0.0010 - 346ms/epoch - 7ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.02619
48/48 - 0s - loss: 0.0122 - mse: 0.0122 - mae: 0.0879 - val_loss: 0.0262 - val_mse: 0.0262 - val_mae: 0.1425 - lr: 0.0010 - 330ms/epoch - 7ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.02619 to 0.01865, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0134 - mse: 0.0134 - mae: 0.0908 - val_loss: 0.0187 - val_mse: 0.0187 - val_mae: 0.1158 - lr: 0.0010 - 322ms/epoch - 7ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.01865
48/48 - 0s - loss: 0.0126 - mse: 0.0126 - mae: 0.0902 - val_loss: 0.0296 - val_mse: 0.0296 - val_mae: 0.1537 - lr: 0.0010 - 333ms/epoch - 7ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01865
48/48 - 0s - loss: 0.0135 - mse: 0.0135 - mae: 0.0915 - val_loss: 0.0223 - val_mse: 0.0223 - val_mae: 0.1294 - lr: 0.0010 - 315ms/epoch - 7ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01865
48/48 - 0s - loss: 0.0117 - mse: 0.0117 - mae: 0.0853 - val_loss: 0.0217 - val_mse: 0.0217 - val_mae: 0.1271 - lr: 0.0010 - 323ms/epoch - 7ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.01865
48/48 - 0s - loss: 0.0126 - mse: 0.0126 - mae: 0.0885 - val_loss: 0.0242 - val_mse: 0.0242 - val_mae: 0.1365 - lr: 0.0010 - 336ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00015: val_loss did not improve from 0.01865
48/48 - 0s - loss: 0.0129 - mse: 0.0129 - mae: 0.0892 - val_loss: 0.0219 - val_mse: 0.0219 - val_mae: 0.1280 - lr: 0.0010 - 396ms/epoch - 8ms/step
Epoch 16/500
Epoch 00016: val_loss improved from 0.01865 to 0.01294, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0235 - mse: 0.0235 - mae: 0.1254 - val_loss: 0.0129 - val_mse: 0.0129 - val_mae: 0.0940 - lr: 1.0000e-04 - 325ms/epoch - 7ms/step
Epoch 17/500
Epoch 00017: val_loss improved from 0.01294 to 0.01156, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0077 - mse: 0.0077 - mae: 0.0724 - val_loss: 0.0116 - val_mse: 0.0116 - val_mae: 0.0883 - lr: 1.0000e-04 - 353ms/epoch - 7ms/step
Epoch 18/500
Epoch 00018: val_loss improved from 0.01156 to 0.01119, saving model to LSTM3.h5
48/48 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0673 - val_loss: 0.0112 - val_mse: 0.0112 - val_mae: 0.0868 - lr: 1.0000e-04 - 346ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0651 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0873 - lr: 1.0000e-04 - 300ms/epoch - 6ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0643 - val_loss: 0.0115 - val_mse: 0.0115 - val_mae: 0.0883 - lr: 1.0000e-04 - 334ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0636 - val_loss: 0.0116 - val_mse: 0.0116 - val_mae: 0.0891 - lr: 1.0000e-04 - 329ms/epoch - 7ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0646 - val_loss: 0.0118 - val_mse: 0.0118 - val_mae: 0.0898 - lr: 1.0000e-04 - 322ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00023: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0612 - val_loss: 0.0124 - val_mse: 0.0124 - val_mae: 0.0925 - lr: 1.0000e-04 - 320ms/epoch - 7ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0587 - val_loss: 0.0124 - val_mse: 0.0124 - val_mae: 0.0925 - lr: 1.0000e-05 - 401ms/epoch - 8ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0579 - val_loss: 0.0123 - val_mse: 0.0123 - val_mae: 0.0922 - lr: 1.0000e-05 - 314ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0596 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0920 - lr: 1.0000e-05 - 334ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0592 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0917 - lr: 1.0000e-05 - 311ms/epoch - 6ms/step
Epoch 28/500
Epoch 00028: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00028: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0604 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0919 - lr: 1.0000e-05 - 316ms/epoch - 7ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0594 - val_loss: 0.0121 - val_mse: 0.0121 - val_mae: 0.0916 - lr: 1.0000e-05 - 343ms/epoch - 7ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0595 - val_loss: 0.0120 - val_mse: 0.0120 - val_mae: 0.0912 - lr: 1.0000e-05 - 311ms/epoch - 6ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0588 - val_loss: 0.0120 - val_mse: 0.0120 - val_mae: 0.0910 - lr: 1.0000e-05 - 310ms/epoch - 6ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0566 - val_loss: 0.0120 - val_mse: 0.0120 - val_mae: 0.0911 - lr: 1.0000e-05 - 319ms/epoch - 7ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0590 - val_loss: 0.0121 - val_mse: 0.0121 - val_mae: 0.0913 - lr: 1.0000e-05 - 302ms/epoch - 6ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0594 - val_loss: 0.0121 - val_mse: 0.0121 - val_mae: 0.0914 - lr: 1.0000e-05 - 333ms/epoch - 7ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0561 - val_loss: 0.0121 - val_mse: 0.0121 - val_mae: 0.0916 - lr: 1.0000e-05 - 313ms/epoch - 7ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0575 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0919 - lr: 1.0000e-05 - 335ms/epoch - 7ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0603 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0920 - lr: 1.0000e-05 - 309ms/epoch - 6ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0594 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0919 - lr: 1.0000e-05 - 340ms/epoch - 7ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0580 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0918 - lr: 1.0000e-05 - 319ms/epoch - 7ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0565 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0920 - lr: 1.0000e-05 - 327ms/epoch - 7ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0586 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0922 - lr: 1.0000e-05 - 346ms/epoch - 7ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0574 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0922 - lr: 1.0000e-05 - 329ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0603 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0921 - lr: 1.0000e-05 - 327ms/epoch - 7ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0607 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0918 - lr: 1.0000e-05 - 314ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0587 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0919 - lr: 1.0000e-05 - 320ms/epoch - 7ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0586 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0920 - lr: 1.0000e-05 - 309ms/epoch - 6ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0559 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0920 - lr: 1.0000e-05 - 327ms/epoch - 7ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0569 - val_loss: 0.0123 - val_mse: 0.0123 - val_mae: 0.0923 - lr: 1.0000e-05 - 350ms/epoch - 7ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0555 - val_loss: 0.0123 - val_mse: 0.0123 - val_mae: 0.0923 - lr: 1.0000e-05 - 340ms/epoch - 7ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0561 - val_loss: 0.0124 - val_mse: 0.0124 - val_mae: 0.0928 - lr: 1.0000e-05 - 322ms/epoch - 7ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0588 - val_loss: 0.0125 - val_mse: 0.0125 - val_mae: 0.0935 - lr: 1.0000e-05 - 322ms/epoch - 7ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0587 - val_loss: 0.0126 - val_mse: 0.0126 - val_mae: 0.0937 - lr: 1.0000e-05 - 306ms/epoch - 6ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0576 - val_loss: 0.0126 - val_mse: 0.0126 - val_mae: 0.0939 - lr: 1.0000e-05 - 316ms/epoch - 7ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0576 - val_loss: 0.0127 - val_mse: 0.0127 - val_mae: 0.0941 - lr: 1.0000e-05 - 427ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0551 - val_loss: 0.0127 - val_mse: 0.0127 - val_mae: 0.0942 - lr: 1.0000e-05 - 320ms/epoch - 7ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0561 - val_loss: 0.0127 - val_mse: 0.0127 - val_mae: 0.0941 - lr: 1.0000e-05 - 314ms/epoch - 7ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0571 - val_loss: 0.0127 - val_mse: 0.0127 - val_mae: 0.0943 - lr: 1.0000e-05 - 317ms/epoch - 7ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0047 - mse: 0.0047 - mae: 0.0553 - val_loss: 0.0127 - val_mse: 0.0127 - val_mae: 0.0943 - lr: 1.0000e-05 - 309ms/epoch - 6ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0573 - val_loss: 0.0128 - val_mse: 0.0128 - val_mae: 0.0946 - lr: 1.0000e-05 - 307ms/epoch - 6ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0548 - val_loss: 0.0128 - val_mse: 0.0128 - val_mae: 0.0948 - lr: 1.0000e-05 - 331ms/epoch - 7ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0554 - val_loss: 0.0128 - val_mse: 0.0128 - val_mae: 0.0948 - lr: 1.0000e-05 - 317ms/epoch - 7ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0050 - mse: 0.0050 - mae: 0.0560 - val_loss: 0.0129 - val_mse: 0.0129 - val_mae: 0.0952 - lr: 1.0000e-05 - 301ms/epoch - 6ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0561 - val_loss: 0.0130 - val_mse: 0.0130 - val_mae: 0.0953 - lr: 1.0000e-05 - 327ms/epoch - 7ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0531 - val_loss: 0.0129 - val_mse: 0.0129 - val_mae: 0.0951 - lr: 1.0000e-05 - 306ms/epoch - 6ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0047 - mse: 0.0047 - mae: 0.0551 - val_loss: 0.0129 - val_mse: 0.0129 - val_mae: 0.0950 - lr: 1.0000e-05 - 325ms/epoch - 7ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0547 - val_loss: 0.0130 - val_mse: 0.0130 - val_mae: 0.0955 - lr: 1.0000e-05 - 321ms/epoch - 7ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0575 - val_loss: 0.0130 - val_mse: 0.0130 - val_mae: 0.0956 - lr: 1.0000e-05 - 311ms/epoch - 6ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.01119
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0560 - val_loss: 0.0131 - val_mse: 0.0131 - val_mae: 0.0958 - lr: 1.0000e-05 - 310ms/epoch - 6ms/step
Epoch 00068: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 50.75% Accuracy
MSE: 60.72793966434877
RMSE: 7.792813334370892
MAPE: 6.245110563496496
EMA
EMA([input_arrays], [timeperiod=30])
Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
51
Working on EMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.59 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4231.556, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3761.238, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.39 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3532.227, Time=0.11 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3394.496, Time=0.13 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.13 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.87 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3396.496, Time=0.26 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 3.590 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1693.248
Date: Sun, 12 Dec 2021 AIC 3394.496
Time: 12:42:57 BIC 3413.260
Sample: 0 HQIC 3401.702
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1982 0.003 -389.569 0.000 -1.204 -1.192
ar.L2 -0.8976 0.006 -139.811 0.000 -0.910 -0.885
ar.L3 -0.3984 0.006 -68.662 0.000 -0.410 -0.387
sigma2 3.9230 0.018 215.372 0.000 3.887 3.959
===================================================================================
Ljung-Box (L1) (Q): 14.54 Jarque-Bera (JB): 2462173.05
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 273.82
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.39802, saving model to LSTM3.h5
16/16 - 3s - loss: 0.1339 - mse: 0.1339 - mae: 0.3073 - val_loss: 0.3980 - val_mse: 0.3980 - val_mae: 0.6034 - lr: 0.0010 - 3s/epoch - 184ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.39802
16/16 - 0s - loss: 0.0284 - mse: 0.0284 - mae: 0.1365 - val_loss: 0.4921 - val_mse: 0.4921 - val_mae: 0.6766 - lr: 0.0010 - 119ms/epoch - 7ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.39802 to 0.34963, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0339 - mse: 0.0339 - mae: 0.1483 - val_loss: 0.3496 - val_mse: 0.3496 - val_mae: 0.5668 - lr: 0.0010 - 139ms/epoch - 9ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.34963
16/16 - 0s - loss: 0.0149 - mse: 0.0149 - mae: 0.0965 - val_loss: 0.3667 - val_mse: 0.3667 - val_mae: 0.5819 - lr: 0.0010 - 125ms/epoch - 8ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.34963 to 0.31417, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0164 - mse: 0.0164 - mae: 0.1027 - val_loss: 0.3142 - val_mse: 0.3142 - val_mae: 0.5373 - lr: 0.0010 - 201ms/epoch - 13ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.31417
16/16 - 0s - loss: 0.0141 - mse: 0.0141 - mae: 0.0931 - val_loss: 0.3175 - val_mse: 0.3175 - val_mae: 0.5408 - lr: 0.0010 - 126ms/epoch - 8ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.31417 to 0.28925, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0143 - mse: 0.0143 - mae: 0.0942 - val_loss: 0.2892 - val_mse: 0.2892 - val_mae: 0.5153 - lr: 0.0010 - 151ms/epoch - 9ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.28925 to 0.28776, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0103 - mse: 0.0103 - mae: 0.0789 - val_loss: 0.2878 - val_mse: 0.2878 - val_mae: 0.5139 - lr: 0.0010 - 156ms/epoch - 10ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.28776 to 0.26242, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0102 - mse: 0.0102 - mae: 0.0799 - val_loss: 0.2624 - val_mse: 0.2624 - val_mae: 0.4898 - lr: 0.0010 - 174ms/epoch - 11ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.26242 to 0.24907, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0091 - mse: 0.0091 - mae: 0.0749 - val_loss: 0.2491 - val_mse: 0.2491 - val_mae: 0.4767 - lr: 0.0010 - 147ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: val_loss improved from 0.24907 to 0.22973, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0091 - mse: 0.0091 - mae: 0.0767 - val_loss: 0.2297 - val_mse: 0.2297 - val_mae: 0.4568 - lr: 0.0010 - 156ms/epoch - 10ms/step
Epoch 12/500
Epoch 00012: val_loss improved from 0.22973 to 0.22391, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0089 - mse: 0.0089 - mae: 0.0745 - val_loss: 0.2239 - val_mse: 0.2239 - val_mae: 0.4507 - lr: 0.0010 - 157ms/epoch - 10ms/step
Epoch 13/500
Epoch 00013: val_loss improved from 0.22391 to 0.20284, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0073 - mse: 0.0073 - mae: 0.0677 - val_loss: 0.2028 - val_mse: 0.2028 - val_mae: 0.4277 - lr: 0.0010 - 198ms/epoch - 12ms/step
Epoch 14/500
Epoch 00014: val_loss improved from 0.20284 to 0.19709, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0660 - val_loss: 0.1971 - val_mse: 0.1971 - val_mae: 0.4211 - lr: 0.0010 - 146ms/epoch - 9ms/step
Epoch 15/500
Epoch 00015: val_loss improved from 0.19709 to 0.18772, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0083 - mse: 0.0083 - mae: 0.0718 - val_loss: 0.1877 - val_mse: 0.1877 - val_mae: 0.4103 - lr: 0.0010 - 153ms/epoch - 10ms/step
Epoch 16/500
Epoch 00016: val_loss improved from 0.18772 to 0.16471, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0082 - mse: 0.0082 - mae: 0.0724 - val_loss: 0.1647 - val_mse: 0.1647 - val_mae: 0.3824 - lr: 0.0010 - 150ms/epoch - 9ms/step
Epoch 17/500
Epoch 00017: val_loss improved from 0.16471 to 0.16153, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0609 - val_loss: 0.1615 - val_mse: 0.1615 - val_mae: 0.3779 - lr: 0.0010 - 164ms/epoch - 10ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.16153
16/16 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0602 - val_loss: 0.1621 - val_mse: 0.1621 - val_mae: 0.3785 - lr: 0.0010 - 138ms/epoch - 9ms/step
Epoch 19/500
Epoch 00019: val_loss improved from 0.16153 to 0.14689, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0652 - val_loss: 0.1469 - val_mse: 0.1469 - val_mae: 0.3588 - lr: 0.0010 - 203ms/epoch - 13ms/step
Epoch 20/500
Epoch 00020: val_loss improved from 0.14689 to 0.13731, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0605 - val_loss: 0.1373 - val_mse: 0.1373 - val_mae: 0.3458 - lr: 0.0010 - 142ms/epoch - 9ms/step
Epoch 21/500
Epoch 00021: val_loss improved from 0.13731 to 0.13575, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0587 - val_loss: 0.1358 - val_mse: 0.1358 - val_mae: 0.3438 - lr: 0.0010 - 152ms/epoch - 9ms/step
Epoch 22/500
Epoch 00022: val_loss improved from 0.13575 to 0.12955, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0635 - val_loss: 0.1296 - val_mse: 0.1296 - val_mae: 0.3351 - lr: 0.0010 - 150ms/epoch - 9ms/step
Epoch 23/500
Epoch 00023: val_loss improved from 0.12955 to 0.11616, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0609 - val_loss: 0.1162 - val_mse: 0.1162 - val_mae: 0.3151 - lr: 0.0010 - 153ms/epoch - 10ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.11616
16/16 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0604 - val_loss: 0.1163 - val_mse: 0.1163 - val_mae: 0.3151 - lr: 0.0010 - 136ms/epoch - 8ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.11616
16/16 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0577 - val_loss: 0.1294 - val_mse: 0.1294 - val_mae: 0.3339 - lr: 0.0010 - 126ms/epoch - 8ms/step
Epoch 26/500
Epoch 00026: val_loss improved from 0.11616 to 0.11583, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0575 - val_loss: 0.1158 - val_mse: 0.1158 - val_mae: 0.3147 - lr: 0.0010 - 156ms/epoch - 10ms/step
Epoch 27/500
Epoch 00027: val_loss improved from 0.11583 to 0.10589, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0574 - val_loss: 0.1059 - val_mse: 0.1059 - val_mae: 0.2999 - lr: 0.0010 - 159ms/epoch - 10ms/step
Epoch 28/500
Epoch 00028: val_loss improved from 0.10589 to 0.10370, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0552 - val_loss: 0.1037 - val_mse: 0.1037 - val_mae: 0.2963 - lr: 0.0010 - 154ms/epoch - 10ms/step
Epoch 29/500
Epoch 00029: val_loss improved from 0.10370 to 0.10333, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0564 - val_loss: 0.1033 - val_mse: 0.1033 - val_mae: 0.2954 - lr: 0.0010 - 148ms/epoch - 9ms/step
Epoch 30/500
Epoch 00030: val_loss improved from 0.10333 to 0.09683, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0593 - val_loss: 0.0968 - val_mse: 0.0968 - val_mae: 0.2844 - lr: 0.0010 - 157ms/epoch - 10ms/step
Epoch 31/500
Epoch 00031: val_loss improved from 0.09683 to 0.08818, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0632 - val_loss: 0.0882 - val_mse: 0.0882 - val_mae: 0.2699 - lr: 0.0010 - 156ms/epoch - 10ms/step
Epoch 32/500
Epoch 00032: val_loss improved from 0.08818 to 0.07657, saving model to LSTM3.h5
16/16 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0578 - val_loss: 0.0766 - val_mse: 0.0766 - val_mae: 0.2495 - lr: 0.0010 - 156ms/epoch - 10ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0518 - val_loss: 0.0859 - val_mse: 0.0859 - val_mae: 0.2660 - lr: 0.0010 - 172ms/epoch - 11ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0535 - val_loss: 0.0876 - val_mse: 0.0876 - val_mae: 0.2690 - lr: 0.0010 - 132ms/epoch - 8ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0050 - mse: 0.0050 - mae: 0.0545 - val_loss: 0.0862 - val_mse: 0.0862 - val_mae: 0.2667 - lr: 0.0010 - 136ms/epoch - 9ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0537 - val_loss: 0.0810 - val_mse: 0.0810 - val_mae: 0.2577 - lr: 0.0010 - 122ms/epoch - 8ms/step
Epoch 37/500
Epoch 00037: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00037: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0553 - val_loss: 0.0869 - val_mse: 0.0869 - val_mae: 0.2678 - lr: 0.0010 - 133ms/epoch - 8ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0517 - val_loss: 0.0832 - val_mse: 0.0832 - val_mae: 0.2615 - lr: 1.0000e-04 - 135ms/epoch - 8ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0047 - mse: 0.0047 - mae: 0.0541 - val_loss: 0.0803 - val_mse: 0.0803 - val_mae: 0.2565 - lr: 1.0000e-04 - 132ms/epoch - 8ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0511 - val_loss: 0.0810 - val_mse: 0.0810 - val_mae: 0.2578 - lr: 1.0000e-04 - 135ms/epoch - 8ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0509 - val_loss: 0.0809 - val_mse: 0.0809 - val_mae: 0.2577 - lr: 1.0000e-04 - 128ms/epoch - 8ms/step
Epoch 42/500
Epoch 00042: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00042: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0504 - val_loss: 0.0806 - val_mse: 0.0806 - val_mae: 0.2572 - lr: 1.0000e-04 - 133ms/epoch - 8ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0488 - val_loss: 0.0807 - val_mse: 0.0807 - val_mae: 0.2573 - lr: 1.0000e-05 - 119ms/epoch - 7ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0508 - val_loss: 0.0808 - val_mse: 0.0808 - val_mae: 0.2574 - lr: 1.0000e-05 - 144ms/epoch - 9ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0520 - val_loss: 0.0808 - val_mse: 0.0808 - val_mae: 0.2576 - lr: 1.0000e-05 - 176ms/epoch - 11ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0481 - val_loss: 0.0810 - val_mse: 0.0810 - val_mae: 0.2579 - lr: 1.0000e-05 - 142ms/epoch - 9ms/step
Epoch 47/500
Epoch 00047: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00047: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0483 - val_loss: 0.0812 - val_mse: 0.0812 - val_mae: 0.2583 - lr: 1.0000e-05 - 128ms/epoch - 8ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0497 - val_loss: 0.0813 - val_mse: 0.0813 - val_mae: 0.2584 - lr: 1.0000e-05 - 137ms/epoch - 9ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0500 - val_loss: 0.0813 - val_mse: 0.0813 - val_mae: 0.2584 - lr: 1.0000e-05 - 127ms/epoch - 8ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0482 - val_loss: 0.0813 - val_mse: 0.0813 - val_mae: 0.2584 - lr: 1.0000e-05 - 130ms/epoch - 8ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0504 - val_loss: 0.0813 - val_mse: 0.0813 - val_mae: 0.2584 - lr: 1.0000e-05 - 134ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0487 - val_loss: 0.0814 - val_mse: 0.0814 - val_mae: 0.2585 - lr: 1.0000e-05 - 129ms/epoch - 8ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0491 - val_loss: 0.0813 - val_mse: 0.0813 - val_mae: 0.2584 - lr: 1.0000e-05 - 132ms/epoch - 8ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0505 - val_loss: 0.0814 - val_mse: 0.0814 - val_mae: 0.2586 - lr: 1.0000e-05 - 141ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0479 - val_loss: 0.0814 - val_mse: 0.0814 - val_mae: 0.2585 - lr: 1.0000e-05 - 134ms/epoch - 8ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0500 - val_loss: 0.0812 - val_mse: 0.0812 - val_mae: 0.2583 - lr: 1.0000e-05 - 136ms/epoch - 9ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0483 - val_loss: 0.0813 - val_mse: 0.0813 - val_mae: 0.2584 - lr: 1.0000e-05 - 128ms/epoch - 8ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0035 - mse: 0.0035 - mae: 0.0467 - val_loss: 0.0814 - val_mse: 0.0814 - val_mae: 0.2586 - lr: 1.0000e-05 - 140ms/epoch - 9ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0505 - val_loss: 0.0815 - val_mse: 0.0815 - val_mae: 0.2587 - lr: 1.0000e-05 - 128ms/epoch - 8ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0491 - val_loss: 0.0814 - val_mse: 0.0814 - val_mae: 0.2586 - lr: 1.0000e-05 - 129ms/epoch - 8ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0513 - val_loss: 0.0814 - val_mse: 0.0814 - val_mae: 0.2586 - lr: 1.0000e-05 - 128ms/epoch - 8ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0493 - val_loss: 0.0814 - val_mse: 0.0814 - val_mae: 0.2587 - lr: 1.0000e-05 - 138ms/epoch - 9ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0482 - val_loss: 0.0814 - val_mse: 0.0814 - val_mae: 0.2587 - lr: 1.0000e-05 - 139ms/epoch - 9ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0477 - val_loss: 0.0814 - val_mse: 0.0814 - val_mae: 0.2586 - lr: 1.0000e-05 - 124ms/epoch - 8ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0509 - val_loss: 0.0813 - val_mse: 0.0813 - val_mae: 0.2584 - lr: 1.0000e-05 - 146ms/epoch - 9ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0487 - val_loss: 0.0812 - val_mse: 0.0812 - val_mae: 0.2584 - lr: 1.0000e-05 - 130ms/epoch - 8ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0475 - val_loss: 0.0812 - val_mse: 0.0812 - val_mae: 0.2583 - lr: 1.0000e-05 - 122ms/epoch - 8ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0500 - val_loss: 0.0811 - val_mse: 0.0811 - val_mae: 0.2581 - lr: 1.0000e-05 - 131ms/epoch - 8ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0516 - val_loss: 0.0811 - val_mse: 0.0811 - val_mae: 0.2581 - lr: 1.0000e-05 - 129ms/epoch - 8ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0500 - val_loss: 0.0810 - val_mse: 0.0810 - val_mae: 0.2580 - lr: 1.0000e-05 - 124ms/epoch - 8ms/step
Epoch 71/500
Epoch 00071: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0513 - val_loss: 0.0808 - val_mse: 0.0808 - val_mae: 0.2576 - lr: 1.0000e-05 - 131ms/epoch - 8ms/step
Epoch 72/500
Epoch 00072: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0515 - val_loss: 0.0807 - val_mse: 0.0807 - val_mae: 0.2575 - lr: 1.0000e-05 - 125ms/epoch - 8ms/step
Epoch 73/500
Epoch 00073: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0482 - val_loss: 0.0808 - val_mse: 0.0808 - val_mae: 0.2576 - lr: 1.0000e-05 - 151ms/epoch - 9ms/step
Epoch 74/500
Epoch 00074: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0505 - val_loss: 0.0808 - val_mse: 0.0808 - val_mae: 0.2577 - lr: 1.0000e-05 - 131ms/epoch - 8ms/step
Epoch 75/500
Epoch 00075: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0495 - val_loss: 0.0807 - val_mse: 0.0807 - val_mae: 0.2575 - lr: 1.0000e-05 - 133ms/epoch - 8ms/step
Epoch 76/500
Epoch 00076: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0509 - val_loss: 0.0806 - val_mse: 0.0806 - val_mae: 0.2574 - lr: 1.0000e-05 - 128ms/epoch - 8ms/step
Epoch 77/500
Epoch 00077: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0502 - val_loss: 0.0807 - val_mse: 0.0807 - val_mae: 0.2574 - lr: 1.0000e-05 - 130ms/epoch - 8ms/step
Epoch 78/500
Epoch 00078: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0488 - val_loss: 0.0807 - val_mse: 0.0807 - val_mae: 0.2575 - lr: 1.0000e-05 - 130ms/epoch - 8ms/step
Epoch 79/500
Epoch 00079: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0490 - val_loss: 0.0807 - val_mse: 0.0807 - val_mae: 0.2575 - lr: 1.0000e-05 - 130ms/epoch - 8ms/step
Epoch 80/500
Epoch 00080: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0490 - val_loss: 0.0806 - val_mse: 0.0806 - val_mae: 0.2572 - lr: 1.0000e-05 - 197ms/epoch - 12ms/step
Epoch 81/500
Epoch 00081: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0510 - val_loss: 0.0805 - val_mse: 0.0805 - val_mae: 0.2572 - lr: 1.0000e-05 - 130ms/epoch - 8ms/step
Epoch 82/500
Epoch 00082: val_loss did not improve from 0.07657
16/16 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0500 - val_loss: 0.0803 - val_mse: 0.0803 - val_mae: 0.2569 - lr: 1.0000e-05 - 132ms/epoch - 8ms/step
Epoch 00082: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 50.75% Accuracy
MSE: 60.72793966434877
RMSE: 7.792813334370892
MAPE: 6.245110563496496
EMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 24.31424046870049
RMSE: 4.930947218202654
MAPE: 4.072073008160127
WMA
WMA([input_arrays], [timeperiod=30])
Weighted Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
49
Working on WMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.59 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4264.089, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3793.930, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.32 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3564.923, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3427.258, Time=0.12 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.70 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.64 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3429.258, Time=0.24 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 3.813 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1709.629
Date: Sun, 12 Dec 2021 AIC 3427.258
Time: 12:44:50 BIC 3446.021
Sample: 0 HQIC 3434.464
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1981 0.003 -389.386 0.000 -1.204 -1.192
ar.L2 -0.8974 0.006 -139.699 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.737 0.000 -0.410 -0.387
sigma2 4.0860 0.019 215.311 0.000 4.049 4.123
===================================================================================
Ljung-Box (L1) (Q): 14.57 Jarque-Bera (JB): 2460901.70
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 273.75
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.20576, saving model to LSTM3.h5
17/17 - 3s - loss: 0.1644 - mse: 0.1644 - mae: 0.2939 - val_loss: 0.2058 - val_mse: 0.2058 - val_mae: 0.4444 - lr: 0.0010 - 3s/epoch - 179ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.20576 to 0.12386, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0361 - mse: 0.0361 - mae: 0.1547 - val_loss: 0.1239 - val_mse: 0.1239 - val_mae: 0.3397 - lr: 0.0010 - 169ms/epoch - 10ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.12386 to 0.09635, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0239 - mse: 0.0239 - mae: 0.1202 - val_loss: 0.0964 - val_mse: 0.0964 - val_mae: 0.2956 - lr: 0.0010 - 164ms/epoch - 10ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.09635 to 0.06144, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0161 - mse: 0.0161 - mae: 0.0996 - val_loss: 0.0614 - val_mse: 0.0614 - val_mae: 0.2271 - lr: 0.0010 - 151ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.06144 to 0.04145, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0152 - mse: 0.0152 - mae: 0.0969 - val_loss: 0.0414 - val_mse: 0.0414 - val_mae: 0.1773 - lr: 0.0010 - 150ms/epoch - 9ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.04145 to 0.02859, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0118 - mse: 0.0118 - mae: 0.0858 - val_loss: 0.0286 - val_mse: 0.0286 - val_mae: 0.1427 - lr: 0.0010 - 159ms/epoch - 9ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.02859 to 0.02033, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0119 - mse: 0.0119 - mae: 0.0860 - val_loss: 0.0203 - val_mse: 0.0203 - val_mae: 0.1198 - lr: 0.0010 - 201ms/epoch - 12ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.02033 to 0.01659, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0098 - mse: 0.0098 - mae: 0.0780 - val_loss: 0.0166 - val_mse: 0.0166 - val_mae: 0.1087 - lr: 0.0010 - 157ms/epoch - 9ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.01659 to 0.01485, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0091 - mse: 0.0091 - mae: 0.0749 - val_loss: 0.0148 - val_mse: 0.0148 - val_mae: 0.1034 - lr: 0.0010 - 167ms/epoch - 10ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.01485 to 0.01360, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0086 - mse: 0.0086 - mae: 0.0728 - val_loss: 0.0136 - val_mse: 0.0136 - val_mae: 0.0995 - lr: 0.0010 - 158ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: val_loss improved from 0.01360 to 0.01305, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0089 - mse: 0.0089 - mae: 0.0735 - val_loss: 0.0131 - val_mse: 0.0131 - val_mae: 0.0975 - lr: 0.0010 - 155ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss improved from 0.01305 to 0.01222, saving model to LSTM3.h5
17/17 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0654 - val_loss: 0.0122 - val_mse: 0.0122 - val_mae: 0.0946 - lr: 0.0010 - 164ms/epoch - 10ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0675 - val_loss: 0.0123 - val_mse: 0.0123 - val_mae: 0.0940 - lr: 0.0010 - 135ms/epoch - 8ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0071 - mse: 0.0071 - mae: 0.0667 - val_loss: 0.0123 - val_mse: 0.0123 - val_mae: 0.0933 - lr: 0.0010 - 135ms/epoch - 8ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0078 - mse: 0.0078 - mae: 0.0675 - val_loss: 0.0128 - val_mse: 0.0128 - val_mae: 0.0934 - lr: 0.0010 - 173ms/epoch - 10ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0662 - val_loss: 0.0136 - val_mse: 0.0136 - val_mae: 0.0943 - lr: 0.0010 - 140ms/epoch - 8ms/step
Epoch 17/500
Epoch 00017: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00017: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0595 - val_loss: 0.0160 - val_mse: 0.0160 - val_mae: 0.0998 - lr: 0.0010 - 153ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0623 - val_loss: 0.0157 - val_mse: 0.0157 - val_mae: 0.0991 - lr: 1.0000e-04 - 148ms/epoch - 9ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0588 - val_loss: 0.0155 - val_mse: 0.0155 - val_mae: 0.0984 - lr: 1.0000e-04 - 135ms/epoch - 8ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0601 - val_loss: 0.0153 - val_mse: 0.0153 - val_mae: 0.0980 - lr: 1.0000e-04 - 188ms/epoch - 11ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0620 - val_loss: 0.0152 - val_mse: 0.0152 - val_mae: 0.0975 - lr: 1.0000e-04 - 178ms/epoch - 10ms/step
Epoch 22/500
Epoch 00022: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00022: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0608 - val_loss: 0.0150 - val_mse: 0.0150 - val_mae: 0.0969 - lr: 1.0000e-04 - 141ms/epoch - 8ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0638 - val_loss: 0.0150 - val_mse: 0.0150 - val_mae: 0.0970 - lr: 1.0000e-05 - 133ms/epoch - 8ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0582 - val_loss: 0.0151 - val_mse: 0.0151 - val_mae: 0.0971 - lr: 1.0000e-05 - 139ms/epoch - 8ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0619 - val_loss: 0.0151 - val_mse: 0.0151 - val_mae: 0.0972 - lr: 1.0000e-05 - 140ms/epoch - 8ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0627 - val_loss: 0.0151 - val_mse: 0.0151 - val_mae: 0.0973 - lr: 1.0000e-05 - 138ms/epoch - 8ms/step
Epoch 27/500
Epoch 00027: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00027: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0595 - val_loss: 0.0152 - val_mse: 0.0152 - val_mae: 0.0973 - lr: 1.0000e-05 - 146ms/epoch - 9ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0577 - val_loss: 0.0152 - val_mse: 0.0152 - val_mae: 0.0974 - lr: 1.0000e-05 - 193ms/epoch - 11ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0596 - val_loss: 0.0152 - val_mse: 0.0152 - val_mae: 0.0973 - lr: 1.0000e-05 - 171ms/epoch - 10ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0594 - val_loss: 0.0152 - val_mse: 0.0152 - val_mae: 0.0974 - lr: 1.0000e-05 - 139ms/epoch - 8ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0576 - val_loss: 0.0152 - val_mse: 0.0152 - val_mae: 0.0975 - lr: 1.0000e-05 - 150ms/epoch - 9ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0593 - val_loss: 0.0152 - val_mse: 0.0152 - val_mae: 0.0975 - lr: 1.0000e-05 - 139ms/epoch - 8ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0593 - val_loss: 0.0153 - val_mse: 0.0153 - val_mae: 0.0976 - lr: 1.0000e-05 - 136ms/epoch - 8ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0593 - val_loss: 0.0153 - val_mse: 0.0153 - val_mae: 0.0978 - lr: 1.0000e-05 - 155ms/epoch - 9ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0615 - val_loss: 0.0154 - val_mse: 0.0154 - val_mae: 0.0978 - lr: 1.0000e-05 - 146ms/epoch - 9ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0592 - val_loss: 0.0154 - val_mse: 0.0154 - val_mae: 0.0979 - lr: 1.0000e-05 - 128ms/epoch - 8ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0589 - val_loss: 0.0155 - val_mse: 0.0155 - val_mae: 0.0980 - lr: 1.0000e-05 - 140ms/epoch - 8ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0562 - val_loss: 0.0155 - val_mse: 0.0155 - val_mae: 0.0980 - lr: 1.0000e-05 - 140ms/epoch - 8ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0604 - val_loss: 0.0155 - val_mse: 0.0155 - val_mae: 0.0981 - lr: 1.0000e-05 - 138ms/epoch - 8ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0622 - val_loss: 0.0155 - val_mse: 0.0155 - val_mae: 0.0981 - lr: 1.0000e-05 - 142ms/epoch - 8ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0594 - val_loss: 0.0155 - val_mse: 0.0155 - val_mae: 0.0981 - lr: 1.0000e-05 - 148ms/epoch - 9ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0578 - val_loss: 0.0155 - val_mse: 0.0155 - val_mae: 0.0982 - lr: 1.0000e-05 - 131ms/epoch - 8ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0604 - val_loss: 0.0155 - val_mse: 0.0155 - val_mae: 0.0982 - lr: 1.0000e-05 - 142ms/epoch - 8ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0569 - val_loss: 0.0156 - val_mse: 0.0156 - val_mae: 0.0983 - lr: 1.0000e-05 - 143ms/epoch - 8ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0579 - val_loss: 0.0157 - val_mse: 0.0157 - val_mae: 0.0984 - lr: 1.0000e-05 - 134ms/epoch - 8ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0568 - val_loss: 0.0157 - val_mse: 0.0157 - val_mae: 0.0984 - lr: 1.0000e-05 - 146ms/epoch - 9ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0560 - val_loss: 0.0157 - val_mse: 0.0157 - val_mae: 0.0984 - lr: 1.0000e-05 - 134ms/epoch - 8ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0598 - val_loss: 0.0157 - val_mse: 0.0157 - val_mae: 0.0985 - lr: 1.0000e-05 - 146ms/epoch - 9ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0594 - val_loss: 0.0157 - val_mse: 0.0157 - val_mae: 0.0984 - lr: 1.0000e-05 - 133ms/epoch - 8ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0593 - val_loss: 0.0157 - val_mse: 0.0157 - val_mae: 0.0985 - lr: 1.0000e-05 - 137ms/epoch - 8ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0577 - val_loss: 0.0158 - val_mse: 0.0158 - val_mae: 0.0986 - lr: 1.0000e-05 - 144ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0584 - val_loss: 0.0158 - val_mse: 0.0158 - val_mae: 0.0987 - lr: 1.0000e-05 - 143ms/epoch - 8ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0566 - val_loss: 0.0158 - val_mse: 0.0158 - val_mae: 0.0987 - lr: 1.0000e-05 - 145ms/epoch - 9ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0563 - val_loss: 0.0158 - val_mse: 0.0158 - val_mae: 0.0987 - lr: 1.0000e-05 - 179ms/epoch - 11ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0579 - val_loss: 0.0159 - val_mse: 0.0159 - val_mae: 0.0989 - lr: 1.0000e-05 - 140ms/epoch - 8ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0554 - val_loss: 0.0160 - val_mse: 0.0160 - val_mae: 0.0991 - lr: 1.0000e-05 - 143ms/epoch - 8ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0567 - val_loss: 0.0160 - val_mse: 0.0160 - val_mae: 0.0992 - lr: 1.0000e-05 - 145ms/epoch - 9ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0589 - val_loss: 0.0160 - val_mse: 0.0160 - val_mae: 0.0992 - lr: 1.0000e-05 - 142ms/epoch - 8ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0553 - val_loss: 0.0160 - val_mse: 0.0160 - val_mae: 0.0992 - lr: 1.0000e-05 - 145ms/epoch - 9ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0597 - val_loss: 0.0161 - val_mse: 0.0161 - val_mae: 0.0993 - lr: 1.0000e-05 - 139ms/epoch - 8ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0569 - val_loss: 0.0161 - val_mse: 0.0161 - val_mae: 0.0994 - lr: 1.0000e-05 - 190ms/epoch - 11ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.01222
17/17 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0599 - val_loss: 0.0161 - val_mse: 0.0161 - val_mae: 0.0994 - lr: 1.0000e-05 - 143ms/epoch - 8ms/step
Epoch 00062: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 50.75% Accuracy
MSE: 60.72793966434877
RMSE: 7.792813334370892
MAPE: 6.245110563496496
EMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 24.31424046870049
RMSE: 4.930947218202654
MAPE: 4.072073008160127
WMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 68.15170187847146
RMSE: 8.255404404296101
MAPE: 6.806281257852447
DEMA
DEMA([input_arrays], [timeperiod=30])
Double Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
89
Working on DEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.58 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4436.126, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3965.317, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.53 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3736.589, Time=0.09 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3598.951, Time=0.11 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.23 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.25 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3600.951, Time=0.27 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.158 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1795.475
Date: Sun, 12 Dec 2021 AIC 3598.951
Time: 12:46:42 BIC 3617.714
Sample: 0 HQIC 3606.157
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1983 0.003 -389.581 0.000 -1.204 -1.192
ar.L2 -0.8973 0.006 -139.732 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.649 0.000 -0.410 -0.387
sigma2 5.0573 0.023 215.292 0.000 5.011 5.103
===================================================================================
Ljung-Box (L1) (Q): 14.41 Jarque-Bera (JB): 2460553.80
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.89
Prob(H) (two-sided): 0.00 Kurtosis: 273.74
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.27877, saving model to LSTM3.h5
10/10 - 3s - loss: 0.7021 - mse: 0.7021 - mae: 0.7244 - val_loss: 0.2788 - val_mse: 0.2788 - val_mae: 0.4616 - lr: 0.0010 - 3s/epoch - 297ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.27877 to 0.16710, saving model to LSTM3.h5
10/10 - 0s - loss: 0.0997 - mse: 0.0997 - mae: 0.2588 - val_loss: 0.1671 - val_mse: 0.1671 - val_mae: 0.3443 - lr: 0.0010 - 117ms/epoch - 12ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.16710 to 0.13466, saving model to LSTM3.h5
10/10 - 0s - loss: 0.0339 - mse: 0.0339 - mae: 0.1506 - val_loss: 0.1347 - val_mse: 0.1347 - val_mae: 0.3038 - lr: 0.0010 - 114ms/epoch - 11ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0374 - mse: 0.0374 - mae: 0.1632 - val_loss: 0.1425 - val_mse: 0.1425 - val_mae: 0.3146 - lr: 0.0010 - 93ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0234 - mse: 0.0234 - mae: 0.1232 - val_loss: 0.1594 - val_mse: 0.1594 - val_mae: 0.3359 - lr: 0.0010 - 97ms/epoch - 10ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0177 - mse: 0.0177 - mae: 0.1065 - val_loss: 0.1713 - val_mse: 0.1713 - val_mae: 0.3504 - lr: 0.0010 - 100ms/epoch - 10ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0197 - mse: 0.0197 - mae: 0.1111 - val_loss: 0.1758 - val_mse: 0.1758 - val_mae: 0.3559 - lr: 0.0010 - 96ms/epoch - 10ms/step
Epoch 8/500
Epoch 00008: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00008: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0170 - mse: 0.0170 - mae: 0.1021 - val_loss: 0.1782 - val_mse: 0.1782 - val_mae: 0.3590 - lr: 0.0010 - 95ms/epoch - 10ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.1004 - val_loss: 0.1781 - val_mse: 0.1781 - val_mae: 0.3590 - lr: 1.0000e-04 - 103ms/epoch - 10ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0166 - mse: 0.0166 - mae: 0.1016 - val_loss: 0.1778 - val_mse: 0.1778 - val_mae: 0.3586 - lr: 1.0000e-04 - 102ms/epoch - 10ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0164 - mse: 0.0164 - mae: 0.1015 - val_loss: 0.1776 - val_mse: 0.1776 - val_mae: 0.3583 - lr: 1.0000e-04 - 91ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0164 - mse: 0.0164 - mae: 0.1014 - val_loss: 0.1773 - val_mse: 0.1773 - val_mae: 0.3580 - lr: 1.0000e-04 - 98ms/epoch - 10ms/step
Epoch 13/500
Epoch 00013: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00013: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0157 - mse: 0.0157 - mae: 0.0999 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3581 - lr: 1.0000e-04 - 95ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0149 - mse: 0.0149 - mae: 0.0967 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3581 - lr: 1.0000e-05 - 94ms/epoch - 9ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0153 - mse: 0.0153 - mae: 0.0972 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3581 - lr: 1.0000e-05 - 98ms/epoch - 10ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0157 - mse: 0.0157 - mae: 0.0994 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3581 - lr: 1.0000e-05 - 104ms/epoch - 10ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.0982 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 134ms/epoch - 13ms/step
Epoch 18/500
Epoch 00018: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00018: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0138 - mse: 0.0138 - mae: 0.0920 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3582 - lr: 1.0000e-05 - 100ms/epoch - 10ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0157 - mse: 0.0157 - mae: 0.0978 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 95ms/epoch - 10ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0154 - mse: 0.0154 - mae: 0.0989 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3582 - lr: 1.0000e-05 - 96ms/epoch - 10ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0163 - mse: 0.0163 - mae: 0.1008 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 95ms/epoch - 10ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0164 - mse: 0.0164 - mae: 0.1016 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 87ms/epoch - 9ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0153 - mse: 0.0153 - mae: 0.0977 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 90ms/epoch - 9ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0169 - mse: 0.0169 - mae: 0.1015 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 96ms/epoch - 10ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0159 - mse: 0.0159 - mae: 0.1013 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 92ms/epoch - 9ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0152 - mse: 0.0152 - mae: 0.0978 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 89ms/epoch - 9ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.0991 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3581 - lr: 1.0000e-05 - 92ms/epoch - 9ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0153 - mse: 0.0153 - mae: 0.0991 - val_loss: 0.1773 - val_mse: 0.1773 - val_mae: 0.3581 - lr: 1.0000e-05 - 92ms/epoch - 9ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0157 - mse: 0.0157 - mae: 0.0995 - val_loss: 0.1773 - val_mse: 0.1773 - val_mae: 0.3581 - lr: 1.0000e-05 - 97ms/epoch - 10ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0973 - val_loss: 0.1773 - val_mse: 0.1773 - val_mae: 0.3581 - lr: 1.0000e-05 - 100ms/epoch - 10ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0145 - mse: 0.0145 - mae: 0.0945 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3581 - lr: 1.0000e-05 - 96ms/epoch - 10ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0154 - mse: 0.0154 - mae: 0.0973 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 136ms/epoch - 14ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0175 - mse: 0.0175 - mae: 0.1044 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 99ms/epoch - 10ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0151 - mse: 0.0151 - mae: 0.0962 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3583 - lr: 1.0000e-05 - 132ms/epoch - 13ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0164 - mse: 0.0164 - mae: 0.0994 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3583 - lr: 1.0000e-05 - 94ms/epoch - 9ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0152 - mse: 0.0152 - mae: 0.0971 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3583 - lr: 1.0000e-05 - 91ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0950 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3583 - lr: 1.0000e-05 - 112ms/epoch - 11ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0953 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 101ms/epoch - 10ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0147 - mse: 0.0147 - mae: 0.0948 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 104ms/epoch - 10ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0150 - mse: 0.0150 - mae: 0.0976 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 95ms/epoch - 9ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0151 - mse: 0.0151 - mae: 0.0961 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 100ms/epoch - 10ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0163 - mse: 0.0163 - mae: 0.1008 - val_loss: 0.1774 - val_mse: 0.1774 - val_mae: 0.3582 - lr: 1.0000e-05 - 140ms/epoch - 14ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0161 - mse: 0.0161 - mae: 0.0997 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3583 - lr: 1.0000e-05 - 96ms/epoch - 10ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0145 - mse: 0.0145 - mae: 0.0941 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3584 - lr: 1.0000e-05 - 98ms/epoch - 10ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.0993 - val_loss: 0.1776 - val_mse: 0.1776 - val_mae: 0.3584 - lr: 1.0000e-05 - 104ms/epoch - 10ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0161 - mse: 0.0161 - mae: 0.0998 - val_loss: 0.1776 - val_mse: 0.1776 - val_mae: 0.3585 - lr: 1.0000e-05 - 115ms/epoch - 11ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0153 - mse: 0.0153 - mae: 0.0972 - val_loss: 0.1776 - val_mse: 0.1776 - val_mae: 0.3585 - lr: 1.0000e-05 - 111ms/epoch - 11ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.0990 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3584 - lr: 1.0000e-05 - 112ms/epoch - 11ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0147 - mse: 0.0147 - mae: 0.0938 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3584 - lr: 1.0000e-05 - 101ms/epoch - 10ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0155 - mse: 0.0155 - mae: 0.0990 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3584 - lr: 1.0000e-05 - 138ms/epoch - 14ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0149 - mse: 0.0149 - mae: 0.0980 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3584 - lr: 1.0000e-05 - 94ms/epoch - 9ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0140 - mse: 0.0140 - mae: 0.0938 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3584 - lr: 1.0000e-05 - 136ms/epoch - 14ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.13466
10/10 - 0s - loss: 0.0145 - mse: 0.0145 - mae: 0.0969 - val_loss: 0.1775 - val_mse: 0.1775 - val_mae: 0.3584 - lr: 1.0000e-05 - 98ms/epoch - 10ms/step
Epoch 00053: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 50.75% Accuracy
MSE: 60.72793966434877
RMSE: 7.792813334370892
MAPE: 6.245110563496496
EMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 24.31424046870049
RMSE: 4.930947218202654
MAPE: 4.072073008160127
WMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 68.15170187847146
RMSE: 8.255404404296101
MAPE: 6.806281257852447
DEMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.51% Accuracy
MSE: 43.922177447330554
RMSE: 6.627380888958364
MAPE: 5.414540694927783
KAMA
KAMA([input_arrays], [timeperiod=30])
Kaufman Adaptive Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
18
Working on KAMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.49 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4190.464, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3724.371, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.38 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3494.154, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3357.435, Time=0.14 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.57 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.98 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3359.435, Time=0.28 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.053 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1674.717
Date: Sun, 12 Dec 2021 AIC 3357.435
Time: 12:48:18 BIC 3376.198
Sample: 0 HQIC 3364.641
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1955 0.003 -381.246 0.000 -1.202 -1.189
ar.L2 -0.8964 0.007 -135.835 0.000 -0.909 -0.883
ar.L3 -0.3971 0.006 -67.229 0.000 -0.409 -0.385
sigma2 3.7466 0.018 211.623 0.000 3.712 3.781
===================================================================================
Ljung-Box (L1) (Q): 14.20 Jarque-Bera (JB): 2338363.32
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.01 Skew: 3.76
Prob(H) (two-sided): 0.00 Kurtosis: 266.93
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.39067, saving model to LSTM3.h5
45/45 - 3s - loss: 0.1630 - mse: 0.1630 - mae: 0.3178 - val_loss: 0.3907 - val_mse: 0.3907 - val_mae: 0.5781 - lr: 0.0010 - 3s/epoch - 71ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.39067 to 0.11867, saving model to LSTM3.h5
45/45 - 0s - loss: 0.0775 - mse: 0.0775 - mae: 0.2131 - val_loss: 0.1187 - val_mse: 0.1187 - val_mae: 0.2877 - lr: 0.0010 - 311ms/epoch - 7ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.11867 to 0.09234, saving model to LSTM3.h5
45/45 - 0s - loss: 0.0253 - mse: 0.0253 - mae: 0.1252 - val_loss: 0.0923 - val_mse: 0.0923 - val_mae: 0.2457 - lr: 0.0010 - 330ms/epoch - 7ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0167 - mse: 0.0167 - mae: 0.1023 - val_loss: 0.1103 - val_mse: 0.1103 - val_mae: 0.2789 - lr: 0.0010 - 308ms/epoch - 7ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0149 - mse: 0.0149 - mae: 0.0963 - val_loss: 0.1277 - val_mse: 0.1277 - val_mae: 0.3071 - lr: 0.0010 - 303ms/epoch - 7ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0150 - mse: 0.0150 - mae: 0.0949 - val_loss: 0.1309 - val_mse: 0.1309 - val_mae: 0.3139 - lr: 0.0010 - 324ms/epoch - 7ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0141 - mse: 0.0141 - mae: 0.0921 - val_loss: 0.1362 - val_mse: 0.1362 - val_mae: 0.3233 - lr: 0.0010 - 324ms/epoch - 7ms/step
Epoch 8/500
Epoch 00008: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00008: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0126 - mse: 0.0126 - mae: 0.0863 - val_loss: 0.1368 - val_mse: 0.1368 - val_mae: 0.3251 - lr: 0.0010 - 336ms/epoch - 7ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0213 - mse: 0.0213 - mae: 0.1182 - val_loss: 0.1069 - val_mse: 0.1069 - val_mae: 0.2808 - lr: 1.0000e-04 - 342ms/epoch - 8ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0084 - mse: 0.0084 - mae: 0.0737 - val_loss: 0.1024 - val_mse: 0.1024 - val_mae: 0.2733 - lr: 1.0000e-04 - 300ms/epoch - 7ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0662 - val_loss: 0.1019 - val_mse: 0.1019 - val_mae: 0.2722 - lr: 1.0000e-04 - 311ms/epoch - 7ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0654 - val_loss: 0.1024 - val_mse: 0.1024 - val_mae: 0.2729 - lr: 1.0000e-04 - 302ms/epoch - 7ms/step
Epoch 13/500
Epoch 00013: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00013: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0691 - val_loss: 0.1030 - val_mse: 0.1030 - val_mae: 0.2738 - lr: 1.0000e-04 - 292ms/epoch - 6ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0662 - val_loss: 0.1030 - val_mse: 0.1030 - val_mae: 0.2738 - lr: 1.0000e-05 - 305ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0625 - val_loss: 0.1031 - val_mse: 0.1031 - val_mae: 0.2739 - lr: 1.0000e-05 - 315ms/epoch - 7ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0627 - val_loss: 0.1030 - val_mse: 0.1030 - val_mae: 0.2737 - lr: 1.0000e-05 - 296ms/epoch - 7ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0675 - val_loss: 0.1029 - val_mse: 0.1029 - val_mae: 0.2735 - lr: 1.0000e-05 - 327ms/epoch - 7ms/step
Epoch 18/500
Epoch 00018: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00018: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0628 - val_loss: 0.1029 - val_mse: 0.1029 - val_mae: 0.2736 - lr: 1.0000e-05 - 318ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0648 - val_loss: 0.1029 - val_mse: 0.1029 - val_mae: 0.2735 - lr: 1.0000e-05 - 306ms/epoch - 7ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0652 - val_loss: 0.1029 - val_mse: 0.1029 - val_mae: 0.2735 - lr: 1.0000e-05 - 303ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0659 - val_loss: 0.1032 - val_mse: 0.1032 - val_mae: 0.2739 - lr: 1.0000e-05 - 297ms/epoch - 7ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0638 - val_loss: 0.1033 - val_mse: 0.1033 - val_mae: 0.2741 - lr: 1.0000e-05 - 320ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0640 - val_loss: 0.1034 - val_mse: 0.1034 - val_mae: 0.2743 - lr: 1.0000e-05 - 325ms/epoch - 7ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0643 - val_loss: 0.1033 - val_mse: 0.1033 - val_mae: 0.2741 - lr: 1.0000e-05 - 307ms/epoch - 7ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0634 - val_loss: 0.1035 - val_mse: 0.1035 - val_mae: 0.2744 - lr: 1.0000e-05 - 318ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0630 - val_loss: 0.1039 - val_mse: 0.1039 - val_mae: 0.2751 - lr: 1.0000e-05 - 324ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0646 - val_loss: 0.1042 - val_mse: 0.1042 - val_mae: 0.2755 - lr: 1.0000e-05 - 292ms/epoch - 6ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0647 - val_loss: 0.1045 - val_mse: 0.1045 - val_mae: 0.2759 - lr: 1.0000e-05 - 329ms/epoch - 7ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0633 - val_loss: 0.1044 - val_mse: 0.1044 - val_mae: 0.2758 - lr: 1.0000e-05 - 301ms/epoch - 7ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0615 - val_loss: 0.1045 - val_mse: 0.1045 - val_mae: 0.2760 - lr: 1.0000e-05 - 297ms/epoch - 7ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0654 - val_loss: 0.1049 - val_mse: 0.1049 - val_mae: 0.2766 - lr: 1.0000e-05 - 323ms/epoch - 7ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0625 - val_loss: 0.1053 - val_mse: 0.1053 - val_mae: 0.2773 - lr: 1.0000e-05 - 314ms/epoch - 7ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0608 - val_loss: 0.1054 - val_mse: 0.1054 - val_mae: 0.2774 - lr: 1.0000e-05 - 310ms/epoch - 7ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0618 - val_loss: 0.1056 - val_mse: 0.1056 - val_mae: 0.2777 - lr: 1.0000e-05 - 305ms/epoch - 7ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0616 - val_loss: 0.1057 - val_mse: 0.1057 - val_mae: 0.2778 - lr: 1.0000e-05 - 310ms/epoch - 7ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0661 - val_loss: 0.1058 - val_mse: 0.1058 - val_mae: 0.2781 - lr: 1.0000e-05 - 310ms/epoch - 7ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0616 - val_loss: 0.1059 - val_mse: 0.1059 - val_mae: 0.2781 - lr: 1.0000e-05 - 312ms/epoch - 7ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0615 - val_loss: 0.1061 - val_mse: 0.1061 - val_mae: 0.2785 - lr: 1.0000e-05 - 302ms/epoch - 7ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0617 - val_loss: 0.1063 - val_mse: 0.1063 - val_mae: 0.2788 - lr: 1.0000e-05 - 317ms/epoch - 7ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0635 - val_loss: 0.1066 - val_mse: 0.1066 - val_mae: 0.2792 - lr: 1.0000e-05 - 293ms/epoch - 7ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0600 - val_loss: 0.1067 - val_mse: 0.1067 - val_mae: 0.2794 - lr: 1.0000e-05 - 335ms/epoch - 7ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0603 - val_loss: 0.1066 - val_mse: 0.1066 - val_mae: 0.2793 - lr: 1.0000e-05 - 307ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0608 - val_loss: 0.1067 - val_mse: 0.1067 - val_mae: 0.2793 - lr: 1.0000e-05 - 315ms/epoch - 7ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0605 - val_loss: 0.1067 - val_mse: 0.1067 - val_mae: 0.2793 - lr: 1.0000e-05 - 318ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0625 - val_loss: 0.1068 - val_mse: 0.1068 - val_mae: 0.2794 - lr: 1.0000e-05 - 323ms/epoch - 7ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0647 - val_loss: 0.1069 - val_mse: 0.1069 - val_mae: 0.2797 - lr: 1.0000e-05 - 323ms/epoch - 7ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0621 - val_loss: 0.1074 - val_mse: 0.1074 - val_mae: 0.2805 - lr: 1.0000e-05 - 311ms/epoch - 7ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0633 - val_loss: 0.1079 - val_mse: 0.1079 - val_mae: 0.2812 - lr: 1.0000e-05 - 314ms/epoch - 7ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0603 - val_loss: 0.1082 - val_mse: 0.1082 - val_mae: 0.2817 - lr: 1.0000e-05 - 321ms/epoch - 7ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0611 - val_loss: 0.1087 - val_mse: 0.1087 - val_mae: 0.2825 - lr: 1.0000e-05 - 320ms/epoch - 7ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0627 - val_loss: 0.1088 - val_mse: 0.1088 - val_mae: 0.2826 - lr: 1.0000e-05 - 312ms/epoch - 7ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0626 - val_loss: 0.1090 - val_mse: 0.1090 - val_mae: 0.2830 - lr: 1.0000e-05 - 334ms/epoch - 7ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.09234
45/45 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0614 - val_loss: 0.1092 - val_mse: 0.1092 - val_mae: 0.2833 - lr: 1.0000e-05 - 306ms/epoch - 7ms/step
Epoch 00053: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 50.75% Accuracy
MSE: 60.72793966434877
RMSE: 7.792813334370892
MAPE: 6.245110563496496
EMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 24.31424046870049
RMSE: 4.930947218202654
MAPE: 4.072073008160127
WMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 68.15170187847146
RMSE: 8.255404404296101
MAPE: 6.806281257852447
DEMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.51% Accuracy
MSE: 43.922177447330554
RMSE: 6.627380888958364
MAPE: 5.414540694927783
KAMA
Prediction vs Close: 56.34% Accuracy
Prediction vs Prediction: 49.25% Accuracy
MSE: 23.99643218488633
RMSE: 4.8986153334270215
MAPE: 3.8674202618000764
MIDPOINT
MIDPOINT([input_arrays], [timeperiod=14])
MidPoint over period (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 14
Outputs:
real
14
Working on MIDPOINT predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.51 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4212.289, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3747.746, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.33 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3523.401, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3387.759, Time=0.13 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.65 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.15 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3389.758, Time=0.26 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.247 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1689.879
Date: Sun, 12 Dec 2021 AIC 3387.759
Time: 12:50:11 BIC 3406.522
Sample: 0 HQIC 3394.964
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1878 0.003 -345.315 0.000 -1.195 -1.181
ar.L2 -0.8876 0.007 -121.809 0.000 -0.902 -0.873
ar.L3 -0.3957 0.007 -60.127 0.000 -0.409 -0.383
sigma2 3.8904 0.020 193.404 0.000 3.851 3.930
===================================================================================
Ljung-Box (L1) (Q): 13.21 Jarque-Bera (JB): 1659080.01
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.08 Skew: 3.28
Prob(H) (two-sided): 0.00 Kurtosis: 225.31
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.29190, saving model to LSTM3.h5
58/58 - 3s - loss: 0.2526 - mse: 0.2526 - mae: 0.3822 - val_loss: 0.2919 - val_mse: 0.2919 - val_mae: 0.4836 - lr: 0.0010 - 3s/epoch - 58ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.29190 to 0.09505, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0489 - mse: 0.0489 - mae: 0.1754 - val_loss: 0.0950 - val_mse: 0.0950 - val_mae: 0.2416 - lr: 0.0010 - 415ms/epoch - 7ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.09505 to 0.05068, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0288 - mse: 0.0288 - mae: 0.1352 - val_loss: 0.0507 - val_mse: 0.0507 - val_mae: 0.1643 - lr: 0.0010 - 408ms/epoch - 7ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.05068 to 0.03884, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0198 - mse: 0.0198 - mae: 0.1134 - val_loss: 0.0388 - val_mse: 0.0388 - val_mae: 0.1437 - lr: 0.0010 - 395ms/epoch - 7ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.03884 to 0.03849, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0147 - mse: 0.0147 - mae: 0.0983 - val_loss: 0.0385 - val_mse: 0.0385 - val_mae: 0.1416 - lr: 0.0010 - 413ms/epoch - 7ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.03849
58/58 - 0s - loss: 0.0126 - mse: 0.0126 - mae: 0.0897 - val_loss: 0.0468 - val_mse: 0.0468 - val_mae: 0.1582 - lr: 0.0010 - 366ms/epoch - 6ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.03849
58/58 - 0s - loss: 0.0106 - mse: 0.0106 - mae: 0.0803 - val_loss: 0.0513 - val_mse: 0.0513 - val_mae: 0.1701 - lr: 0.0010 - 389ms/epoch - 7ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.03849
58/58 - 0s - loss: 0.0103 - mse: 0.0103 - mae: 0.0789 - val_loss: 0.0522 - val_mse: 0.0522 - val_mae: 0.1758 - lr: 0.0010 - 368ms/epoch - 6ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.03849
58/58 - 0s - loss: 0.0087 - mse: 0.0087 - mae: 0.0724 - val_loss: 0.0560 - val_mse: 0.0560 - val_mae: 0.1883 - lr: 0.0010 - 382ms/epoch - 7ms/step
Epoch 10/500
Epoch 00010: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00010: val_loss did not improve from 0.03849
58/58 - 0s - loss: 0.0092 - mse: 0.0092 - mae: 0.0755 - val_loss: 0.0545 - val_mse: 0.0545 - val_mae: 0.1871 - lr: 0.0010 - 387ms/epoch - 7ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.03849
58/58 - 0s - loss: 0.0151 - mse: 0.0151 - mae: 0.1014 - val_loss: 0.0397 - val_mse: 0.0397 - val_mae: 0.1493 - lr: 1.0000e-04 - 358ms/epoch - 6ms/step
Epoch 12/500
Epoch 00012: val_loss improved from 0.03849 to 0.03841, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0082 - mse: 0.0082 - mae: 0.0727 - val_loss: 0.0384 - val_mse: 0.0384 - val_mae: 0.1459 - lr: 1.0000e-04 - 422ms/epoch - 7ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.03841
58/58 - 0s - loss: 0.0077 - mse: 0.0077 - mae: 0.0706 - val_loss: 0.0389 - val_mse: 0.0389 - val_mae: 0.1471 - lr: 1.0000e-04 - 382ms/epoch - 7ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.03841
58/58 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0675 - val_loss: 0.0387 - val_mse: 0.0387 - val_mae: 0.1464 - lr: 1.0000e-04 - 394ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00015: val_loss did not improve from 0.03841
58/58 - 0s - loss: 0.0073 - mse: 0.0073 - mae: 0.0672 - val_loss: 0.0390 - val_mse: 0.0390 - val_mae: 0.1470 - lr: 1.0000e-04 - 376ms/epoch - 6ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.03841
58/58 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0645 - val_loss: 0.0388 - val_mse: 0.0388 - val_mae: 0.1464 - lr: 1.0000e-05 - 375ms/epoch - 6ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.03841
58/58 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0634 - val_loss: 0.0386 - val_mse: 0.0386 - val_mae: 0.1460 - lr: 1.0000e-05 - 383ms/epoch - 7ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.03841
58/58 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0653 - val_loss: 0.0384 - val_mse: 0.0384 - val_mae: 0.1456 - lr: 1.0000e-05 - 400ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: val_loss improved from 0.03841 to 0.03829, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0658 - val_loss: 0.0383 - val_mse: 0.0383 - val_mae: 0.1452 - lr: 1.0000e-05 - 397ms/epoch - 7ms/step
Epoch 20/500
Epoch 00020: val_loss improved from 0.03829 to 0.03817, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0649 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1449 - lr: 1.0000e-05 - 472ms/epoch - 8ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.03817
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0610 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1449 - lr: 1.0000e-05 - 382ms/epoch - 7ms/step
Epoch 22/500
Epoch 00022: val_loss improved from 0.03817 to 0.03805, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0629 - val_loss: 0.0381 - val_mse: 0.0381 - val_mae: 0.1446 - lr: 1.0000e-05 - 397ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: val_loss improved from 0.03805 to 0.03794, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0642 - val_loss: 0.0379 - val_mse: 0.0379 - val_mae: 0.1443 - lr: 1.0000e-05 - 412ms/epoch - 7ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.03794
58/58 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0640 - val_loss: 0.0380 - val_mse: 0.0380 - val_mae: 0.1446 - lr: 1.0000e-05 - 391ms/epoch - 7ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.03794
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0619 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1450 - lr: 1.0000e-05 - 392ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.03794
58/58 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0641 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1451 - lr: 1.0000e-05 - 389ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.03794
58/58 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0628 - val_loss: 0.0383 - val_mse: 0.0383 - val_mae: 0.1452 - lr: 1.0000e-05 - 380ms/epoch - 7ms/step
Epoch 28/500
Epoch 00028: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00028: val_loss did not improve from 0.03794
58/58 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0649 - val_loss: 0.0384 - val_mse: 0.0384 - val_mae: 0.1454 - lr: 1.0000e-05 - 390ms/epoch - 7ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.03794
58/58 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0661 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1451 - lr: 1.0000e-05 - 393ms/epoch - 7ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.03794
58/58 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0635 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1449 - lr: 1.0000e-05 - 459ms/epoch - 8ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.03794
58/58 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0653 - val_loss: 0.0380 - val_mse: 0.0380 - val_mae: 0.1444 - lr: 1.0000e-05 - 453ms/epoch - 8ms/step
Epoch 32/500
Epoch 00032: val_loss improved from 0.03794 to 0.03787, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0611 - val_loss: 0.0379 - val_mse: 0.0379 - val_mae: 0.1442 - lr: 1.0000e-05 - 402ms/epoch - 7ms/step
Epoch 33/500
Epoch 00033: val_loss improved from 0.03787 to 0.03773, saving model to LSTM3.h5
58/58 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0655 - val_loss: 0.0377 - val_mse: 0.0377 - val_mae: 0.1438 - lr: 1.0000e-05 - 394ms/epoch - 7ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0641 - val_loss: 0.0378 - val_mse: 0.0378 - val_mae: 0.1441 - lr: 1.0000e-05 - 380ms/epoch - 7ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0607 - val_loss: 0.0378 - val_mse: 0.0378 - val_mae: 0.1439 - lr: 1.0000e-05 - 368ms/epoch - 6ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0622 - val_loss: 0.0380 - val_mse: 0.0380 - val_mae: 0.1444 - lr: 1.0000e-05 - 382ms/epoch - 7ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0618 - val_loss: 0.0383 - val_mse: 0.0383 - val_mae: 0.1452 - lr: 1.0000e-05 - 403ms/epoch - 7ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0617 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1449 - lr: 1.0000e-05 - 377ms/epoch - 6ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0618 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1451 - lr: 1.0000e-05 - 375ms/epoch - 6ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0599 - val_loss: 0.0384 - val_mse: 0.0384 - val_mae: 0.1456 - lr: 1.0000e-05 - 456ms/epoch - 8ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0615 - val_loss: 0.0385 - val_mse: 0.0385 - val_mae: 0.1458 - lr: 1.0000e-05 - 382ms/epoch - 7ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0616 - val_loss: 0.0388 - val_mse: 0.0388 - val_mae: 0.1465 - lr: 1.0000e-05 - 382ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0609 - val_loss: 0.0387 - val_mse: 0.0387 - val_mae: 0.1464 - lr: 1.0000e-05 - 400ms/epoch - 7ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0620 - val_loss: 0.0390 - val_mse: 0.0390 - val_mae: 0.1469 - lr: 1.0000e-05 - 378ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0633 - val_loss: 0.0388 - val_mse: 0.0388 - val_mae: 0.1467 - lr: 1.0000e-05 - 379ms/epoch - 7ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0597 - val_loss: 0.0391 - val_mse: 0.0391 - val_mae: 0.1474 - lr: 1.0000e-05 - 374ms/epoch - 6ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0624 - val_loss: 0.0393 - val_mse: 0.0393 - val_mae: 0.1477 - lr: 1.0000e-05 - 386ms/epoch - 7ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0610 - val_loss: 0.0393 - val_mse: 0.0393 - val_mae: 0.1478 - lr: 1.0000e-05 - 457ms/epoch - 8ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0583 - val_loss: 0.0394 - val_mse: 0.0394 - val_mae: 0.1481 - lr: 1.0000e-05 - 385ms/epoch - 7ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0621 - val_loss: 0.0393 - val_mse: 0.0393 - val_mae: 0.1479 - lr: 1.0000e-05 - 459ms/epoch - 8ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0618 - val_loss: 0.0392 - val_mse: 0.0392 - val_mae: 0.1477 - lr: 1.0000e-05 - 379ms/epoch - 7ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0618 - val_loss: 0.0392 - val_mse: 0.0392 - val_mae: 0.1476 - lr: 1.0000e-05 - 380ms/epoch - 7ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0621 - val_loss: 0.0391 - val_mse: 0.0391 - val_mae: 0.1473 - lr: 1.0000e-05 - 392ms/epoch - 7ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0606 - val_loss: 0.0392 - val_mse: 0.0392 - val_mae: 0.1477 - lr: 1.0000e-05 - 382ms/epoch - 7ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0589 - val_loss: 0.0393 - val_mse: 0.0393 - val_mae: 0.1479 - lr: 1.0000e-05 - 368ms/epoch - 6ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0611 - val_loss: 0.0394 - val_mse: 0.0394 - val_mae: 0.1480 - lr: 1.0000e-05 - 395ms/epoch - 7ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0599 - val_loss: 0.0394 - val_mse: 0.0394 - val_mae: 0.1481 - lr: 1.0000e-05 - 372ms/epoch - 6ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0596 - val_loss: 0.0393 - val_mse: 0.0393 - val_mae: 0.1479 - lr: 1.0000e-05 - 379ms/epoch - 7ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0593 - val_loss: 0.0395 - val_mse: 0.0395 - val_mae: 0.1483 - lr: 1.0000e-05 - 381ms/epoch - 7ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0584 - val_loss: 0.0399 - val_mse: 0.0399 - val_mae: 0.1492 - lr: 1.0000e-05 - 374ms/epoch - 6ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0607 - val_loss: 0.0400 - val_mse: 0.0400 - val_mae: 0.1496 - lr: 1.0000e-05 - 394ms/epoch - 7ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0606 - val_loss: 0.0400 - val_mse: 0.0400 - val_mae: 0.1495 - lr: 1.0000e-05 - 377ms/epoch - 7ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0603 - val_loss: 0.0401 - val_mse: 0.0401 - val_mae: 0.1499 - lr: 1.0000e-05 - 382ms/epoch - 7ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0593 - val_loss: 0.0404 - val_mse: 0.0404 - val_mae: 0.1505 - lr: 1.0000e-05 - 376ms/epoch - 6ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0633 - val_loss: 0.0403 - val_mse: 0.0403 - val_mae: 0.1503 - lr: 1.0000e-05 - 384ms/epoch - 7ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0609 - val_loss: 0.0403 - val_mse: 0.0403 - val_mae: 0.1503 - lr: 1.0000e-05 - 394ms/epoch - 7ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0591 - val_loss: 0.0404 - val_mse: 0.0404 - val_mae: 0.1506 - lr: 1.0000e-05 - 366ms/epoch - 6ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0597 - val_loss: 0.0407 - val_mse: 0.0407 - val_mae: 0.1514 - lr: 1.0000e-05 - 376ms/epoch - 6ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0562 - val_loss: 0.0409 - val_mse: 0.0409 - val_mae: 0.1517 - lr: 1.0000e-05 - 372ms/epoch - 6ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0595 - val_loss: 0.0409 - val_mse: 0.0409 - val_mae: 0.1519 - lr: 1.0000e-05 - 367ms/epoch - 6ms/step
Epoch 71/500
Epoch 00071: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0588 - val_loss: 0.0410 - val_mse: 0.0410 - val_mae: 0.1522 - lr: 1.0000e-05 - 390ms/epoch - 7ms/step
Epoch 72/500
Epoch 00072: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0610 - val_loss: 0.0414 - val_mse: 0.0414 - val_mae: 0.1529 - lr: 1.0000e-05 - 379ms/epoch - 7ms/step
Epoch 73/500
Epoch 00073: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0614 - val_loss: 0.0414 - val_mse: 0.0414 - val_mae: 0.1531 - lr: 1.0000e-05 - 380ms/epoch - 7ms/step
Epoch 74/500
Epoch 00074: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0568 - val_loss: 0.0414 - val_mse: 0.0414 - val_mae: 0.1531 - lr: 1.0000e-05 - 399ms/epoch - 7ms/step
Epoch 75/500
Epoch 00075: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0593 - val_loss: 0.0415 - val_mse: 0.0415 - val_mae: 0.1534 - lr: 1.0000e-05 - 369ms/epoch - 6ms/step
Epoch 76/500
Epoch 00076: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0583 - val_loss: 0.0417 - val_mse: 0.0417 - val_mae: 0.1538 - lr: 1.0000e-05 - 393ms/epoch - 7ms/step
Epoch 77/500
Epoch 00077: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0576 - val_loss: 0.0420 - val_mse: 0.0420 - val_mae: 0.1544 - lr: 1.0000e-05 - 381ms/epoch - 7ms/step
Epoch 78/500
Epoch 00078: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0585 - val_loss: 0.0419 - val_mse: 0.0419 - val_mae: 0.1543 - lr: 1.0000e-05 - 361ms/epoch - 6ms/step
Epoch 79/500
Epoch 00079: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0588 - val_loss: 0.0420 - val_mse: 0.0420 - val_mae: 0.1545 - lr: 1.0000e-05 - 385ms/epoch - 7ms/step
Epoch 80/500
Epoch 00080: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0581 - val_loss: 0.0420 - val_mse: 0.0420 - val_mae: 0.1544 - lr: 1.0000e-05 - 367ms/epoch - 6ms/step
Epoch 81/500
Epoch 00081: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0606 - val_loss: 0.0421 - val_mse: 0.0421 - val_mae: 0.1547 - lr: 1.0000e-05 - 390ms/epoch - 7ms/step
Epoch 82/500
Epoch 00082: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0569 - val_loss: 0.0427 - val_mse: 0.0427 - val_mae: 0.1561 - lr: 1.0000e-05 - 374ms/epoch - 6ms/step
Epoch 83/500
Epoch 00083: val_loss did not improve from 0.03773
58/58 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0584 - val_loss: 0.0429 - val_mse: 0.0429 - val_mae: 0.1566 - lr: 1.0000e-05 - 381ms/epoch - 7ms/step
Epoch 00083: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 50.75% Accuracy
MSE: 60.72793966434877
RMSE: 7.792813334370892
MAPE: 6.245110563496496
EMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 24.31424046870049
RMSE: 4.930947218202654
MAPE: 4.072073008160127
WMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 68.15170187847146
RMSE: 8.255404404296101
MAPE: 6.806281257852447
DEMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.51% Accuracy
MSE: 43.922177447330554
RMSE: 6.627380888958364
MAPE: 5.414540694927783
KAMA
Prediction vs Close: 56.34% Accuracy
Prediction vs Prediction: 49.25% Accuracy
MSE: 23.99643218488633
RMSE: 4.8986153334270215
MAPE: 3.8674202618000764
MIDPOINT
Prediction vs Close: 50.37% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 28.7191654730941
RMSE: 5.359026541555295
MAPE: 4.42030651732032
T3
T3([input_arrays], [timeperiod=5], [vfactor=0.7])
Triple Exponential Moving Average (T3) (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 5
vfactor: 0.7
Outputs:
real
19
Working on T3 predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.51 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4414.515, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3944.062, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.51 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3715.173, Time=0.10 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3577.471, Time=0.12 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.83 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.77 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3579.471, Time=0.25 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.208 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1784.736
Date: Sun, 12 Dec 2021 AIC 3577.471
Time: 12:52:18 BIC 3596.235
Sample: 0 HQIC 3584.677
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1982 0.003 -389.844 0.000 -1.204 -1.192
ar.L2 -0.8974 0.006 -139.861 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.862 0.000 -0.410 -0.387
sigma2 4.9242 0.023 215.469 0.000 4.879 4.969
===================================================================================
Ljung-Box (L1) (Q): 14.55 Jarque-Bera (JB): 2468024.38
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 274.15
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.44946, saving model to LSTM3.h5
43/43 - 3s - loss: 0.1358 - mse: 0.1358 - mae: 0.2724 - val_loss: 0.4495 - val_mse: 0.4495 - val_mae: 0.6183 - lr: 0.0010 - 3s/epoch - 77ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.44946 to 0.33516, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0321 - mse: 0.0321 - mae: 0.1409 - val_loss: 0.3352 - val_mse: 0.3352 - val_mae: 0.5241 - lr: 0.0010 - 340ms/epoch - 8ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.33516 to 0.31563, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0187 - mse: 0.0187 - mae: 0.1080 - val_loss: 0.3156 - val_mse: 0.3156 - val_mae: 0.5088 - lr: 0.0010 - 316ms/epoch - 7ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.31563 to 0.28607, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0172 - mse: 0.0172 - mae: 0.1047 - val_loss: 0.2861 - val_mse: 0.2861 - val_mae: 0.4844 - lr: 0.0010 - 318ms/epoch - 7ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.28607 to 0.28490, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0140 - mse: 0.0140 - mae: 0.0930 - val_loss: 0.2849 - val_mse: 0.2849 - val_mae: 0.4851 - lr: 0.0010 - 326ms/epoch - 8ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.28490 to 0.26082, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0154 - mse: 0.0154 - mae: 0.0983 - val_loss: 0.2608 - val_mse: 0.2608 - val_mae: 0.4637 - lr: 0.0010 - 323ms/epoch - 8ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.26082 to 0.21860, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0143 - mse: 0.0143 - mae: 0.0931 - val_loss: 0.2186 - val_mse: 0.2186 - val_mae: 0.4217 - lr: 0.0010 - 325ms/epoch - 8ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.21860 to 0.21581, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0137 - mse: 0.0137 - mae: 0.0916 - val_loss: 0.2158 - val_mse: 0.2158 - val_mae: 0.4206 - lr: 0.0010 - 344ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.21581 to 0.18839, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0141 - mse: 0.0141 - mae: 0.0930 - val_loss: 0.1884 - val_mse: 0.1884 - val_mae: 0.3913 - lr: 0.0010 - 321ms/epoch - 7ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.18839 to 0.17354, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0141 - mse: 0.0141 - mae: 0.0923 - val_loss: 0.1735 - val_mse: 0.1735 - val_mae: 0.3747 - lr: 0.0010 - 319ms/epoch - 7ms/step
Epoch 11/500
Epoch 00011: val_loss improved from 0.17354 to 0.14548, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0163 - mse: 0.0163 - mae: 0.1007 - val_loss: 0.1455 - val_mse: 0.1455 - val_mae: 0.3394 - lr: 0.0010 - 331ms/epoch - 8ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.14548
43/43 - 0s - loss: 0.0140 - mse: 0.0140 - mae: 0.0915 - val_loss: 0.1469 - val_mse: 0.1469 - val_mae: 0.3429 - lr: 0.0010 - 302ms/epoch - 7ms/step
Epoch 13/500
Epoch 00013: val_loss improved from 0.14548 to 0.12886, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.0985 - val_loss: 0.1289 - val_mse: 0.1289 - val_mae: 0.3190 - lr: 0.0010 - 335ms/epoch - 8ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.12886
43/43 - 0s - loss: 0.0170 - mse: 0.0170 - mae: 0.1029 - val_loss: 0.1327 - val_mse: 0.1327 - val_mae: 0.3271 - lr: 0.0010 - 301ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss improved from 0.12886 to 0.11198, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0181 - mse: 0.0181 - mae: 0.1063 - val_loss: 0.1120 - val_mse: 0.1120 - val_mae: 0.2975 - lr: 0.0010 - 317ms/epoch - 7ms/step
Epoch 16/500
Epoch 00016: val_loss improved from 0.11198 to 0.10921, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0172 - mse: 0.0172 - mae: 0.1056 - val_loss: 0.1092 - val_mse: 0.1092 - val_mae: 0.2938 - lr: 0.0010 - 332ms/epoch - 8ms/step
Epoch 17/500
Epoch 00017: val_loss improved from 0.10921 to 0.09012, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0171 - mse: 0.0171 - mae: 0.1050 - val_loss: 0.0901 - val_mse: 0.0901 - val_mae: 0.2618 - lr: 0.0010 - 340ms/epoch - 8ms/step
Epoch 18/500
Epoch 00018: val_loss improved from 0.09012 to 0.08075, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.1023 - val_loss: 0.0807 - val_mse: 0.0807 - val_mae: 0.2465 - lr: 0.0010 - 336ms/epoch - 8ms/step
Epoch 19/500
Epoch 00019: val_loss improved from 0.08075 to 0.07533, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.1039 - val_loss: 0.0753 - val_mse: 0.0753 - val_mae: 0.2365 - lr: 0.0010 - 324ms/epoch - 8ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.07533
43/43 - 0s - loss: 0.0135 - mse: 0.0135 - mae: 0.0955 - val_loss: 0.0876 - val_mse: 0.0876 - val_mae: 0.2597 - lr: 0.0010 - 312ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss improved from 0.07533 to 0.06043, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0119 - mse: 0.0119 - mae: 0.0872 - val_loss: 0.0604 - val_mse: 0.0604 - val_mae: 0.2063 - lr: 0.0010 - 319ms/epoch - 7ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.06043
43/43 - 0s - loss: 0.0103 - mse: 0.0103 - mae: 0.0824 - val_loss: 0.0832 - val_mse: 0.0832 - val_mae: 0.2516 - lr: 0.0010 - 300ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: val_loss improved from 0.06043 to 0.05807, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0087 - mse: 0.0087 - mae: 0.0768 - val_loss: 0.0581 - val_mse: 0.0581 - val_mae: 0.2016 - lr: 0.0010 - 344ms/epoch - 8ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.05807
43/43 - 0s - loss: 0.0087 - mse: 0.0087 - mae: 0.0741 - val_loss: 0.0809 - val_mse: 0.0809 - val_mae: 0.2482 - lr: 0.0010 - 300ms/epoch - 7ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.05807
43/43 - 0s - loss: 0.0081 - mse: 0.0081 - mae: 0.0739 - val_loss: 0.0599 - val_mse: 0.0599 - val_mae: 0.2066 - lr: 0.0010 - 295ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.05807
43/43 - 0s - loss: 0.0085 - mse: 0.0085 - mae: 0.0740 - val_loss: 0.0976 - val_mse: 0.0976 - val_mae: 0.2782 - lr: 0.0010 - 289ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss improved from 0.05807 to 0.04480, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0670 - val_loss: 0.0448 - val_mse: 0.0448 - val_mae: 0.1721 - lr: 0.0010 - 344ms/epoch - 8ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04480
43/43 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0676 - val_loss: 0.0764 - val_mse: 0.0764 - val_mae: 0.2408 - lr: 0.0010 - 312ms/epoch - 7ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04480
43/43 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0654 - val_loss: 0.0476 - val_mse: 0.0476 - val_mae: 0.1803 - lr: 0.0010 - 302ms/epoch - 7ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04480
43/43 - 0s - loss: 0.0076 - mse: 0.0076 - mae: 0.0694 - val_loss: 0.0789 - val_mse: 0.0789 - val_mae: 0.2465 - lr: 0.0010 - 324ms/epoch - 8ms/step
Epoch 31/500
Epoch 00031: val_loss improved from 0.04480 to 0.03793, saving model to LSTM3.h5
43/43 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0646 - val_loss: 0.0379 - val_mse: 0.0379 - val_mae: 0.1565 - lr: 0.0010 - 342ms/epoch - 8ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0688 - val_loss: 0.0968 - val_mse: 0.0968 - val_mae: 0.2792 - lr: 0.0010 - 302ms/epoch - 7ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0638 - val_loss: 0.0428 - val_mse: 0.0428 - val_mae: 0.1703 - lr: 0.0010 - 323ms/epoch - 8ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0637 - val_loss: 0.0938 - val_mse: 0.0938 - val_mae: 0.2747 - lr: 0.0010 - 307ms/epoch - 7ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0640 - val_loss: 0.0404 - val_mse: 0.0404 - val_mae: 0.1647 - lr: 0.0010 - 296ms/epoch - 7ms/step
Epoch 36/500
Epoch 00036: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00036: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0076 - mse: 0.0076 - mae: 0.0661 - val_loss: 0.1039 - val_mse: 0.1039 - val_mae: 0.2926 - lr: 0.0010 - 325ms/epoch - 8ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0095 - mse: 0.0095 - mae: 0.0783 - val_loss: 0.0696 - val_mse: 0.0696 - val_mae: 0.2319 - lr: 1.0000e-04 - 300ms/epoch - 7ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0538 - val_loss: 0.0613 - val_mse: 0.0613 - val_mae: 0.2146 - lr: 1.0000e-04 - 293ms/epoch - 7ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0521 - val_loss: 0.0593 - val_mse: 0.0593 - val_mae: 0.2103 - lr: 1.0000e-04 - 306ms/epoch - 7ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0502 - val_loss: 0.0585 - val_mse: 0.0585 - val_mae: 0.2084 - lr: 1.0000e-04 - 306ms/epoch - 7ms/step
Epoch 41/500
Epoch 00041: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00041: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0497 - val_loss: 0.0577 - val_mse: 0.0577 - val_mae: 0.2067 - lr: 1.0000e-04 - 311ms/epoch - 7ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0505 - val_loss: 0.0573 - val_mse: 0.0573 - val_mae: 0.2058 - lr: 1.0000e-05 - 291ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0503 - val_loss: 0.0571 - val_mse: 0.0571 - val_mae: 0.2053 - lr: 1.0000e-05 - 316ms/epoch - 7ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0500 - val_loss: 0.0571 - val_mse: 0.0571 - val_mae: 0.2053 - lr: 1.0000e-05 - 321ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0485 - val_loss: 0.0569 - val_mse: 0.0569 - val_mae: 0.2048 - lr: 1.0000e-05 - 303ms/epoch - 7ms/step
Epoch 46/500
Epoch 00046: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00046: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0476 - val_loss: 0.0568 - val_mse: 0.0568 - val_mae: 0.2046 - lr: 1.0000e-05 - 317ms/epoch - 7ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0481 - val_loss: 0.0569 - val_mse: 0.0569 - val_mae: 0.2048 - lr: 1.0000e-05 - 337ms/epoch - 8ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0468 - val_loss: 0.0567 - val_mse: 0.0567 - val_mae: 0.2045 - lr: 1.0000e-05 - 303ms/epoch - 7ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0483 - val_loss: 0.0566 - val_mse: 0.0566 - val_mae: 0.2042 - lr: 1.0000e-05 - 306ms/epoch - 7ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0495 - val_loss: 0.0565 - val_mse: 0.0565 - val_mae: 0.2039 - lr: 1.0000e-05 - 313ms/epoch - 7ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0479 - val_loss: 0.0563 - val_mse: 0.0563 - val_mae: 0.2034 - lr: 1.0000e-05 - 292ms/epoch - 7ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0472 - val_loss: 0.0564 - val_mse: 0.0564 - val_mae: 0.2036 - lr: 1.0000e-05 - 313ms/epoch - 7ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0499 - val_loss: 0.0562 - val_mse: 0.0562 - val_mae: 0.2032 - lr: 1.0000e-05 - 289ms/epoch - 7ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0489 - val_loss: 0.0561 - val_mse: 0.0561 - val_mae: 0.2030 - lr: 1.0000e-05 - 319ms/epoch - 7ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0479 - val_loss: 0.0560 - val_mse: 0.0560 - val_mae: 0.2028 - lr: 1.0000e-05 - 319ms/epoch - 7ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0501 - val_loss: 0.0561 - val_mse: 0.0561 - val_mae: 0.2030 - lr: 1.0000e-05 - 316ms/epoch - 7ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0471 - val_loss: 0.0563 - val_mse: 0.0563 - val_mae: 0.2034 - lr: 1.0000e-05 - 312ms/epoch - 7ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0471 - val_loss: 0.0561 - val_mse: 0.0561 - val_mae: 0.2031 - lr: 1.0000e-05 - 302ms/epoch - 7ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0035 - mse: 0.0035 - mae: 0.0468 - val_loss: 0.0563 - val_mse: 0.0563 - val_mae: 0.2035 - lr: 1.0000e-05 - 303ms/epoch - 7ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0035 - mse: 0.0035 - mae: 0.0466 - val_loss: 0.0562 - val_mse: 0.0562 - val_mae: 0.2031 - lr: 1.0000e-05 - 307ms/epoch - 7ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0480 - val_loss: 0.0560 - val_mse: 0.0560 - val_mae: 0.2029 - lr: 1.0000e-05 - 305ms/epoch - 7ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0500 - val_loss: 0.0557 - val_mse: 0.0557 - val_mae: 0.2022 - lr: 1.0000e-05 - 325ms/epoch - 8ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0498 - val_loss: 0.0558 - val_mse: 0.0558 - val_mae: 0.2023 - lr: 1.0000e-05 - 296ms/epoch - 7ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0511 - val_loss: 0.0557 - val_mse: 0.0557 - val_mae: 0.2021 - lr: 1.0000e-05 - 294ms/epoch - 7ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0470 - val_loss: 0.0559 - val_mse: 0.0559 - val_mae: 0.2024 - lr: 1.0000e-05 - 330ms/epoch - 8ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0491 - val_loss: 0.0557 - val_mse: 0.0557 - val_mae: 0.2022 - lr: 1.0000e-05 - 294ms/epoch - 7ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0477 - val_loss: 0.0556 - val_mse: 0.0556 - val_mae: 0.2020 - lr: 1.0000e-05 - 287ms/epoch - 7ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0505 - val_loss: 0.0554 - val_mse: 0.0554 - val_mae: 0.2013 - lr: 1.0000e-05 - 322ms/epoch - 7ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0466 - val_loss: 0.0555 - val_mse: 0.0555 - val_mae: 0.2017 - lr: 1.0000e-05 - 294ms/epoch - 7ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0490 - val_loss: 0.0558 - val_mse: 0.0558 - val_mae: 0.2024 - lr: 1.0000e-05 - 290ms/epoch - 7ms/step
Epoch 71/500
Epoch 00071: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0494 - val_loss: 0.0558 - val_mse: 0.0558 - val_mae: 0.2024 - lr: 1.0000e-05 - 308ms/epoch - 7ms/step
Epoch 72/500
Epoch 00072: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0484 - val_loss: 0.0556 - val_mse: 0.0556 - val_mae: 0.2019 - lr: 1.0000e-05 - 327ms/epoch - 8ms/step
Epoch 73/500
Epoch 00073: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0469 - val_loss: 0.0558 - val_mse: 0.0558 - val_mae: 0.2023 - lr: 1.0000e-05 - 293ms/epoch - 7ms/step
Epoch 74/500
Epoch 00074: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0490 - val_loss: 0.0555 - val_mse: 0.0555 - val_mae: 0.2017 - lr: 1.0000e-05 - 303ms/epoch - 7ms/step
Epoch 75/500
Epoch 00075: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0487 - val_loss: 0.0559 - val_mse: 0.0559 - val_mae: 0.2025 - lr: 1.0000e-05 - 312ms/epoch - 7ms/step
Epoch 76/500
Epoch 00076: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0478 - val_loss: 0.0559 - val_mse: 0.0559 - val_mae: 0.2025 - lr: 1.0000e-05 - 293ms/epoch - 7ms/step
Epoch 77/500
Epoch 00077: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0499 - val_loss: 0.0560 - val_mse: 0.0560 - val_mae: 0.2027 - lr: 1.0000e-05 - 280ms/epoch - 7ms/step
Epoch 78/500
Epoch 00078: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0485 - val_loss: 0.0558 - val_mse: 0.0558 - val_mae: 0.2024 - lr: 1.0000e-05 - 335ms/epoch - 8ms/step
Epoch 79/500
Epoch 00079: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0481 - val_loss: 0.0558 - val_mse: 0.0558 - val_mae: 0.2024 - lr: 1.0000e-05 - 294ms/epoch - 7ms/step
Epoch 80/500
Epoch 00080: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0494 - val_loss: 0.0558 - val_mse: 0.0558 - val_mae: 0.2024 - lr: 1.0000e-05 - 304ms/epoch - 7ms/step
Epoch 81/500
Epoch 00081: val_loss did not improve from 0.03793
43/43 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0490 - val_loss: 0.0559 - val_mse: 0.0559 - val_mae: 0.2027 - lr: 1.0000e-05 - 322ms/epoch - 7ms/step
Epoch 00081: early stopping
SMA
Prediction vs Close: 52.61% Accuracy
Prediction vs Prediction: 50.75% Accuracy
MSE: 60.72793966434877
RMSE: 7.792813334370892
MAPE: 6.245110563496496
EMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 24.31424046870049
RMSE: 4.930947218202654
MAPE: 4.072073008160127
WMA
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 68.15170187847146
RMSE: 8.255404404296101
MAPE: 6.806281257852447
DEMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.51% Accuracy
MSE: 43.922177447330554
RMSE: 6.627380888958364
MAPE: 5.414540694927783
KAMA
Prediction vs Close: 56.34% Accuracy
Prediction vs Prediction: 49.25% Accuracy
MSE: 23.99643218488633
RMSE: 4.8986153334270215
MAPE: 3.8674202618000764
MIDPOINT
Prediction vs Close: 50.37% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 28.7191654730941
RMSE: 5.359026541555295
MAPE: 4.42030651732032
T3
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 47.39% Accuracy
MSE: 57.05690799124099
RMSE: 7.553602318843705
MAPE: 6.064989593585796
TEMA
TEMA([input_arrays], [timeperiod=30])
Triple Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
9
Working on TEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.63 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4352.703, Time=0.04 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3889.412, Time=0.06 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.41 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3689.930, Time=0.08 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3574.245, Time=0.11 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.49 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=1.06 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3576.245, Time=0.28 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 4.162 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1783.123
Date: Sun, 12 Dec 2021 AIC 3574.245
Time: 12:54:26 BIC 3593.008
Sample: 0 HQIC 3581.451
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1480 0.004 -302.430 0.000 -1.155 -1.141
ar.L2 -0.8300 0.008 -99.682 0.000 -0.846 -0.814
ar.L3 -0.3687 0.007 -50.527 0.000 -0.383 -0.354
sigma2 4.9055 0.028 175.970 0.000 4.851 4.960
===================================================================================
Ljung-Box (L1) (Q): 11.61 Jarque-Bera (JB): 1261976.58
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.16 Skew: 2.52
Prob(H) (two-sided): 0.00 Kurtosis: 196.90
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.52336, saving model to LSTM3.h5
90/90 - 4s - loss: 0.1259 - mse: 0.1259 - mae: 0.2372 - val_loss: 0.5234 - val_mse: 0.5234 - val_mae: 0.6793 - lr: 0.0010 - 4s/epoch - 40ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.52336 to 0.46424, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0151 - mse: 0.0151 - mae: 0.0950 - val_loss: 0.4642 - val_mse: 0.4642 - val_mae: 0.6371 - lr: 0.0010 - 601ms/epoch - 7ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.46424 to 0.39030, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0104 - mse: 0.0104 - mae: 0.0811 - val_loss: 0.3903 - val_mse: 0.3903 - val_mae: 0.5815 - lr: 0.0010 - 592ms/epoch - 7ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.39030
90/90 - 1s - loss: 0.0096 - mse: 0.0096 - mae: 0.0776 - val_loss: 0.4059 - val_mse: 0.4059 - val_mae: 0.5945 - lr: 0.0010 - 582ms/epoch - 6ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.39030 to 0.37908, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0107 - mse: 0.0107 - mae: 0.0809 - val_loss: 0.3791 - val_mse: 0.3791 - val_mae: 0.5737 - lr: 0.0010 - 587ms/epoch - 7ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.37908 to 0.34176, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0137 - mse: 0.0137 - mae: 0.0906 - val_loss: 0.3418 - val_mse: 0.3418 - val_mae: 0.5434 - lr: 0.0010 - 593ms/epoch - 7ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.34176
90/90 - 1s - loss: 0.0126 - mse: 0.0126 - mae: 0.0863 - val_loss: 0.3475 - val_mse: 0.3475 - val_mae: 0.5496 - lr: 0.0010 - 590ms/epoch - 7ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.34176 to 0.33384, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0139 - mse: 0.0139 - mae: 0.0918 - val_loss: 0.3338 - val_mse: 0.3338 - val_mae: 0.5411 - lr: 0.0010 - 584ms/epoch - 6ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.33384 to 0.27606, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0103 - mse: 0.0103 - mae: 0.0797 - val_loss: 0.2761 - val_mse: 0.2761 - val_mae: 0.4891 - lr: 0.0010 - 605ms/epoch - 7ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.27606 to 0.27021, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0089 - mse: 0.0089 - mae: 0.0709 - val_loss: 0.2702 - val_mse: 0.2702 - val_mae: 0.4844 - lr: 0.0010 - 594ms/epoch - 7ms/step
Epoch 11/500
Epoch 00011: val_loss improved from 0.27021 to 0.21405, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0061 - mse: 0.0061 - mae: 0.0610 - val_loss: 0.2141 - val_mse: 0.2141 - val_mae: 0.4274 - lr: 0.0010 - 638ms/epoch - 7ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.21405
90/90 - 1s - loss: 0.0064 - mse: 0.0064 - mae: 0.0629 - val_loss: 0.2688 - val_mse: 0.2688 - val_mae: 0.4840 - lr: 0.0010 - 617ms/epoch - 7ms/step
Epoch 13/500
Epoch 00013: val_loss improved from 0.21405 to 0.20815, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0074 - mse: 0.0074 - mae: 0.0658 - val_loss: 0.2081 - val_mse: 0.2081 - val_mae: 0.4220 - lr: 0.0010 - 604ms/epoch - 7ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.20815
90/90 - 1s - loss: 0.0077 - mse: 0.0077 - mae: 0.0658 - val_loss: 0.2552 - val_mse: 0.2552 - val_mae: 0.4721 - lr: 0.0010 - 643ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss improved from 0.20815 to 0.17608, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0070 - mse: 0.0070 - mae: 0.0624 - val_loss: 0.1761 - val_mse: 0.1761 - val_mae: 0.3862 - lr: 0.0010 - 658ms/epoch - 7ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.17608
90/90 - 1s - loss: 0.0076 - mse: 0.0076 - mae: 0.0665 - val_loss: 0.2636 - val_mse: 0.2636 - val_mae: 0.4820 - lr: 0.0010 - 640ms/epoch - 7ms/step
Epoch 17/500
Epoch 00017: val_loss improved from 0.17608 to 0.14463, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0076 - mse: 0.0076 - mae: 0.0641 - val_loss: 0.1446 - val_mse: 0.1446 - val_mae: 0.3483 - lr: 0.0010 - 601ms/epoch - 7ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.14463
90/90 - 1s - loss: 0.0076 - mse: 0.0076 - mae: 0.0638 - val_loss: 0.2627 - val_mse: 0.2627 - val_mae: 0.4835 - lr: 0.0010 - 583ms/epoch - 6ms/step
Epoch 19/500
Epoch 00019: val_loss improved from 0.14463 to 0.10852, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0067 - mse: 0.0067 - mae: 0.0593 - val_loss: 0.1085 - val_mse: 0.1085 - val_mae: 0.2977 - lr: 0.0010 - 656ms/epoch - 7ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.10852
90/90 - 1s - loss: 0.0069 - mse: 0.0069 - mae: 0.0608 - val_loss: 0.2559 - val_mse: 0.2559 - val_mae: 0.4772 - lr: 0.0010 - 598ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss improved from 0.10852 to 0.10372, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0065 - mse: 0.0065 - mae: 0.0597 - val_loss: 0.1037 - val_mse: 0.1037 - val_mae: 0.2916 - lr: 0.0010 - 640ms/epoch - 7ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.10372
90/90 - 1s - loss: 0.0058 - mse: 0.0058 - mae: 0.0590 - val_loss: 0.2050 - val_mse: 0.2050 - val_mae: 0.4258 - lr: 0.0010 - 623ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.10372
90/90 - 1s - loss: 0.0057 - mse: 0.0057 - mae: 0.0566 - val_loss: 0.1106 - val_mse: 0.1106 - val_mae: 0.3051 - lr: 0.0010 - 581ms/epoch - 6ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.10372
90/90 - 1s - loss: 0.0057 - mse: 0.0057 - mae: 0.0583 - val_loss: 0.1798 - val_mse: 0.1798 - val_mae: 0.3982 - lr: 0.0010 - 586ms/epoch - 7ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.10372
90/90 - 1s - loss: 0.0057 - mse: 0.0057 - mae: 0.0577 - val_loss: 0.1084 - val_mse: 0.1084 - val_mae: 0.3016 - lr: 0.0010 - 636ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00026: val_loss did not improve from 0.10372
90/90 - 1s - loss: 0.0060 - mse: 0.0060 - mae: 0.0609 - val_loss: 0.1685 - val_mse: 0.1685 - val_mae: 0.3851 - lr: 0.0010 - 577ms/epoch - 6ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.10372
90/90 - 1s - loss: 0.0107 - mse: 0.0107 - mae: 0.0824 - val_loss: 0.1128 - val_mse: 0.1128 - val_mae: 0.3118 - lr: 1.0000e-04 - 552ms/epoch - 6ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.10372
90/90 - 1s - loss: 0.0045 - mse: 0.0045 - mae: 0.0543 - val_loss: 0.1052 - val_mse: 0.1052 - val_mae: 0.3003 - lr: 1.0000e-04 - 625ms/epoch - 7ms/step
Epoch 29/500
Epoch 00029: val_loss improved from 0.10372 to 0.10207, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0039 - mse: 0.0039 - mae: 0.0494 - val_loss: 0.1021 - val_mse: 0.1021 - val_mae: 0.2952 - lr: 1.0000e-04 - 608ms/epoch - 7ms/step
Epoch 30/500
Epoch 00030: val_loss improved from 0.10207 to 0.10053, saving model to LSTM3.h5
90/90 - 1s - loss: 0.0038 - mse: 0.0038 - mae: 0.0484 - val_loss: 0.1005 - val_mse: 0.1005 - val_mae: 0.2926 - lr: 1.0000e-04 - 575ms/epoch - 6ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0038 - mse: 0.0038 - mae: 0.0502 - val_loss: 0.1011 - val_mse: 0.1011 - val_mae: 0.2933 - lr: 1.0000e-04 - 587ms/epoch - 7ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0036 - mse: 0.0036 - mae: 0.0478 - val_loss: 0.1024 - val_mse: 0.1024 - val_mae: 0.2952 - lr: 1.0000e-04 - 574ms/epoch - 6ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0039 - mse: 0.0039 - mae: 0.0480 - val_loss: 0.1036 - val_mse: 0.1036 - val_mae: 0.2969 - lr: 1.0000e-04 - 580ms/epoch - 6ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0034 - mse: 0.0034 - mae: 0.0462 - val_loss: 0.1085 - val_mse: 0.1085 - val_mae: 0.3043 - lr: 1.0000e-04 - 578ms/epoch - 6ms/step
Epoch 35/500
Epoch 00035: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00035: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0036 - mse: 0.0036 - mae: 0.0464 - val_loss: 0.1101 - val_mse: 0.1101 - val_mae: 0.3066 - lr: 1.0000e-04 - 577ms/epoch - 6ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0035 - mse: 0.0035 - mae: 0.0458 - val_loss: 0.1084 - val_mse: 0.1084 - val_mae: 0.3041 - lr: 1.0000e-05 - 581ms/epoch - 6ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0033 - mse: 0.0033 - mae: 0.0450 - val_loss: 0.1074 - val_mse: 0.1074 - val_mae: 0.3026 - lr: 1.0000e-05 - 633ms/epoch - 7ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0449 - val_loss: 0.1066 - val_mse: 0.1066 - val_mae: 0.3013 - lr: 1.0000e-05 - 577ms/epoch - 6ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0437 - val_loss: 0.1062 - val_mse: 0.1062 - val_mae: 0.3007 - lr: 1.0000e-05 - 590ms/epoch - 7ms/step
Epoch 40/500
Epoch 00040: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00040: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0441 - val_loss: 0.1057 - val_mse: 0.1057 - val_mae: 0.2999 - lr: 1.0000e-05 - 647ms/epoch - 7ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0033 - mse: 0.0033 - mae: 0.0450 - val_loss: 0.1059 - val_mse: 0.1059 - val_mae: 0.3002 - lr: 1.0000e-05 - 569ms/epoch - 6ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0033 - mse: 0.0033 - mae: 0.0459 - val_loss: 0.1061 - val_mse: 0.1061 - val_mae: 0.3006 - lr: 1.0000e-05 - 592ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0030 - mse: 0.0030 - mae: 0.0436 - val_loss: 0.1061 - val_mse: 0.1061 - val_mae: 0.3005 - lr: 1.0000e-05 - 558ms/epoch - 6ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0033 - mse: 0.0033 - mae: 0.0445 - val_loss: 0.1062 - val_mse: 0.1062 - val_mae: 0.3006 - lr: 1.0000e-05 - 597ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0033 - mse: 0.0033 - mae: 0.0440 - val_loss: 0.1061 - val_mse: 0.1061 - val_mae: 0.3004 - lr: 1.0000e-05 - 574ms/epoch - 6ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0034 - mse: 0.0034 - mae: 0.0465 - val_loss: 0.1063 - val_mse: 0.1063 - val_mae: 0.3007 - lr: 1.0000e-05 - 588ms/epoch - 7ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0033 - mse: 0.0033 - mae: 0.0451 - val_loss: 0.1065 - val_mse: 0.1065 - val_mae: 0.3011 - lr: 1.0000e-05 - 641ms/epoch - 7ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0034 - mse: 0.0034 - mae: 0.0458 - val_loss: 0.1064 - val_mse: 0.1064 - val_mae: 0.3009 - lr: 1.0000e-05 - 629ms/epoch - 7ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0444 - val_loss: 0.1060 - val_mse: 0.1060 - val_mae: 0.3002 - lr: 1.0000e-05 - 590ms/epoch - 7ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0031 - mse: 0.0031 - mae: 0.0434 - val_loss: 0.1061 - val_mse: 0.1061 - val_mae: 0.3002 - lr: 1.0000e-05 - 584ms/epoch - 6ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0031 - mse: 0.0031 - mae: 0.0437 - val_loss: 0.1063 - val_mse: 0.1063 - val_mae: 0.3005 - lr: 1.0000e-05 - 587ms/epoch - 7ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0451 - val_loss: 0.1069 - val_mse: 0.1069 - val_mae: 0.3014 - lr: 1.0000e-05 - 618ms/epoch - 7ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0448 - val_loss: 0.1066 - val_mse: 0.1066 - val_mae: 0.3010 - lr: 1.0000e-05 - 561ms/epoch - 6ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0031 - mse: 0.0031 - mae: 0.0429 - val_loss: 0.1068 - val_mse: 0.1068 - val_mae: 0.3012 - lr: 1.0000e-05 - 600ms/epoch - 7ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0441 - val_loss: 0.1076 - val_mse: 0.1076 - val_mae: 0.3024 - lr: 1.0000e-05 - 574ms/epoch - 6ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0033 - mse: 0.0033 - mae: 0.0455 - val_loss: 0.1080 - val_mse: 0.1080 - val_mae: 0.3031 - lr: 1.0000e-05 - 590ms/epoch - 7ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0036 - mse: 0.0036 - mae: 0.0472 - val_loss: 0.1087 - val_mse: 0.1087 - val_mae: 0.3040 - lr: 1.0000e-05 - 604ms/epoch - 7ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0031 - mse: 0.0031 - mae: 0.0450 - val_loss: 0.1091 - val_mse: 0.1091 - val_mae: 0.3045 - lr: 1.0000e-05 - 557ms/epoch - 6ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0440 - val_loss: 0.1087 - val_mse: 0.1087 - val_mae: 0.3039 - lr: 1.0000e-05 - 647ms/epoch - 7ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0027 - mse: 0.0027 - mae: 0.0421 - val_loss: 0.1094 - val_mse: 0.1094 - val_mae: 0.3050 - lr: 1.0000e-05 - 561ms/epoch - 6ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0034 - mse: 0.0034 - mae: 0.0451 - val_loss: 0.1099 - val_mse: 0.1099 - val_mae: 0.3057 - lr: 1.0000e-05 - 566ms/epoch - 6ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0031 - mse: 0.0031 - mae: 0.0438 - val_loss: 0.1107 - val_mse: 0.1107 - val_mae: 0.3068 - lr: 1.0000e-05 - 566ms/epoch - 6ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0031 - mse: 0.0031 - mae: 0.0436 - val_loss: 0.1110 - val_mse: 0.1110 - val_mae: 0.3072 - lr: 1.0000e-05 - 629ms/epoch - 7ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0034 - mse: 0.0034 - mae: 0.0454 - val_loss: 0.1100 - val_mse: 0.1100 - val_mae: 0.3058 - lr: 1.0000e-05 - 577ms/epoch - 6ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0030 - mse: 0.0030 - mae: 0.0434 - val_loss: 0.1098 - val_mse: 0.1098 - val_mae: 0.3055 - lr: 1.0000e-05 - 556ms/epoch - 6ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0030 - mse: 0.0030 - mae: 0.0425 - val_loss: 0.1101 - val_mse: 0.1101 - val_mae: 0.3059 - lr: 1.0000e-05 - 569ms/epoch - 6ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0448 - val_loss: 0.1100 - val_mse: 0.1100 - val_mae: 0.3057 - lr: 1.0000e-05 - 584ms/epoch - 6ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0439 - val_loss: 0.1109 - val_mse: 0.1109 - val_mae: 0.3070 - lr: 1.0000e-05 - 654ms/epoch - 7ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0035 - mse: 0.0035 - mae: 0.0465 - val_loss: 0.1112 - val_mse: 0.1112 - val_mae: 0.3075 - lr: 1.0000e-05 - 638ms/epoch - 7ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0034 - mse: 0.0034 - mae: 0.0460 - val_loss: 0.1107 - val_mse: 0.1107 - val_mae: 0.3066 - lr: 1.0000e-05 - 649ms/epoch - 7ms/step
Epoch 71/500
Epoch 00071: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0034 - mse: 0.0034 - mae: 0.0467 - val_loss: 0.1105 - val_mse: 0.1105 - val_mae: 0.3064 - lr: 1.0000e-05 - 614ms/epoch - 7ms/step
Epoch 72/500
Epoch 00072: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0452 - val_loss: 0.1115 - val_mse: 0.1115 - val_mae: 0.3078 - lr: 1.0000e-05 - 646ms/epoch - 7ms/step
Epoch 73/500
Epoch 00073: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0032 - mse: 0.0032 - mae: 0.0450 - val_loss: 0.1121 - val_mse: 0.1121 - val_mae: 0.3087 - lr: 1.0000e-05 - 577ms/epoch - 6ms/step
Epoch 74/500
Epoch 00074: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0029 - mse: 0.0029 - mae: 0.0429 - val_loss: 0.1126 - val_mse: 0.1126 - val_mae: 0.3093 - lr: 1.0000e-05 - 628ms/epoch - 7ms/step
Epoch 75/500
Epoch 00075: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0031 - mse: 0.0031 - mae: 0.0433 - val_loss: 0.1123 - val_mse: 0.1123 - val_mae: 0.3089 - lr: 1.0000e-05 - 630ms/epoch - 7ms/step
Epoch 76/500
Epoch 00076: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0031 - mse: 0.0031 - mae: 0.0439 - val_loss: 0.1126 - val_mse: 0.1126 - val_mae: 0.3094 - lr: 1.0000e-05 - 593ms/epoch - 7ms/step
Epoch 77/500
Epoch 00077: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0030 - mse: 0.0030 - mae: 0.0424 - val_loss: 0.1139 - val_mse: 0.1139 - val_mae: 0.3112 - lr: 1.0000e-05 - 624ms/epoch - 7ms/step
Epoch 78/500
Epoch 00078: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0030 - mse: 0.0030 - mae: 0.0435 - val_loss: 0.1145 - val_mse: 0.1145 - val_mae: 0.3120 - lr: 1.0000e-05 - 584ms/epoch - 6ms/step
Epoch 79/500
Epoch 00079: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0029 - mse: 0.0029 - mae: 0.0427 - val_loss: 0.1149 - val_mse: 0.1149 - val_mae: 0.3127 - lr: 1.0000e-05 - 627ms/epoch - 7ms/step
Epoch 80/500
Epoch 00080: val_loss did not improve from 0.10053
90/90 - 1s - loss: 0.0033 - mse: 0.0033 - mae: 0.0452 - val_loss: 0.1143 - val_mse: 0.1143 - val_mae: 0.3118 - lr: 1.0000e-05 - 631ms/epoch - 7ms/step
Epoch 00080: early stopping
SMA Prediction vs Close: 52.61% Accuracy Prediction vs Prediction: 50.75% Accuracy MSE: 60.72793966434877 RMSE: 7.792813334370892 MAPE: 6.245110563496496 EMA Prediction vs Close: 53.36% Accuracy Prediction vs Prediction: 49.63% Accuracy MSE: 24.31424046870049 RMSE: 4.930947218202654 MAPE: 4.072073008160127 WMA Prediction vs Close: 53.36% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 68.15170187847146 RMSE: 8.255404404296101 MAPE: 6.806281257852447 DEMA Prediction vs Close: 51.49% Accuracy Prediction vs Prediction: 48.51% Accuracy MSE: 43.922177447330554 RMSE: 6.627380888958364 MAPE: 5.414540694927783 KAMA Prediction vs Close: 56.34% Accuracy Prediction vs Prediction: 49.25% Accuracy MSE: 23.99643218488633 RMSE: 4.8986153334270215 MAPE: 3.8674202618000764 MIDPOINT Prediction vs Close: 50.37% Accuracy Prediction vs Prediction: 48.88% Accuracy MSE: 28.7191654730941 RMSE: 5.359026541555295 MAPE: 4.42030651732032 T3 Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 47.39% Accuracy MSE: 57.05690799124099 RMSE: 7.553602318843705 MAPE: 6.064989593585796 TEMA Prediction vs Close: 48.88% Accuracy Prediction vs Prediction: 48.88% Accuracy MSE: 20.964688340236016 RMSE: 4.578721256009806 MAPE: 3.7212897315589664 Runtime: mins: 16.46426535278334
Architecture Used¶
In [ ]:
from google.colab import files
import cv2
uploaded = files.upload()
Saving Experiment3.png to Experiment3 (1).png
In [ ]:
img = cv2.imread('Experiment3.png')
plt.figure(figsize=(20,10))
plt.axis("off")
plt.title('LSTM Architecture '+imgfile,fontsize=18)
plt.imshow(img)
Out[ ]:
<matplotlib.image.AxesImage at 0x7fdfb2abd6d0>
Model Plots¶
In [75]:
with open('simulation3_data.json') as json_file:
simulation3 = json.load(json_file)
fileimg = 'Experiment3'
In [76]:
for i in range(len(list(simulation3.keys()))):
SIM = list(simulation3.keys())[i]
plot_train(simulation3,SIM)
plot_test(simulation3,SIM)
----- Train RMSE for SMA ----- 8.952567216610497 ----- Train_MSE_LSTM for SMA ----- 80.14845976792903 ----- Train MAE LSTM for SMA ----- 7.821491223024573
----- Test RMSE for SMA----- 7.792813334370892 ----- Test_MSE_LSTM for SMA----- 60.72793966434877 ----- Test_MAE_LSTM for SMA----- 6.245110563496496
----- Train RMSE for EMA ----- 10.570254973011156 ----- Train_MSE_LSTM for EMA ----- 111.73029019446709 ----- Train MAE LSTM for EMA ----- 9.451268005811858
----- Test RMSE for EMA----- 4.930947218202654 ----- Test_MSE_LSTM for EMA----- 24.31424046870049 ----- Test_MAE_LSTM for EMA----- 4.072073008160127
----- Train RMSE for WMA ----- 10.97436711794096 ----- Train_MSE_LSTM for WMA ----- 120.43673363934376 ----- Train MAE LSTM for WMA ----- 9.925815236695067
----- Test RMSE for WMA----- 8.255404404296101 ----- Test_MSE_LSTM for WMA----- 68.15170187847146 ----- Test_MAE_LSTM for WMA----- 6.806281257852447
----- Train RMSE for DEMA ----- 12.066299877592092 ----- Train_MSE_LSTM for DEMA ----- 145.59559273597895 ----- Train MAE LSTM for DEMA ----- 10.907355915123395
----- Test RMSE for DEMA----- 6.627380888958364 ----- Test_MSE_LSTM for DEMA----- 43.922177447330554 ----- Test_MAE_LSTM for DEMA----- 5.414540694927783
----- Train RMSE for KAMA ----- 10.869146791217638 ----- Train_MSE_LSTM for KAMA ----- 118.13835196903668 ----- Train MAE LSTM for KAMA ----- 9.759291692937742
----- Test RMSE for KAMA----- 4.8986153334270215 ----- Test_MSE_LSTM for KAMA----- 23.99643218488633 ----- Test_MAE_LSTM for KAMA----- 3.8674202618000764
----- Train RMSE for MIDPOINT ----- 9.691011765630286 ----- Train_MSE_LSTM for MIDPOINT ----- 93.91570904158462 ----- Train MAE LSTM for MIDPOINT ----- 8.650189046672622
----- Test RMSE for MIDPOINT----- 5.359026541555295 ----- Test_MSE_LSTM for MIDPOINT----- 28.7191654730941 ----- Test_MAE_LSTM for MIDPOINT----- 4.42030651732032
----- Train RMSE for T3 ----- 12.129650565307267 ----- Train_MSE_LSTM for T3 ----- 147.12842283645892 ----- Train MAE LSTM for T3 ----- 10.936634904781638
----- Test RMSE for T3----- 7.553602318843705 ----- Test_MSE_LSTM for T3----- 57.05690799124099 ----- Test_MAE_LSTM for T3----- 6.064989593585796
----- Train RMSE for TEMA ----- 7.376932734913034 ----- Train_MSE_LSTM for TEMA ----- 54.419136575431494 ----- Train MAE LSTM for TEMA ----- 5.050424731629302
----- Test RMSE for TEMA----- 4.578721256009806 ----- Test_MSE_LSTM for TEMA----- 20.964688340236016 ----- Test_MAE_LSTM for TEMA----- 3.7212897315589664
In [ ]:
for i in range(len(list(simulation3.keys()))):
SIM = list(simulation3.keys())[i]
plot_train(simulation3,SIM)
plot_test(simulation3,SIM)
----- Train RMSE for SMA ----- 8.952567216610497 ----- Train_MSE_LSTM for SMA ----- 80.14845976792903 ----- Train MAE LSTM for SMA ----- 7.821491223024573
----- Test RMSE for SMA----- 7.792813334370892 ----- Test_MSE_LSTM for SMA----- 60.72793966434877 ----- Test_MAE_LSTM for SMA----- 6.245110563496496
----- Train RMSE for EMA ----- 10.570254973011156 ----- Train_MSE_LSTM for EMA ----- 111.73029019446709 ----- Train MAE LSTM for EMA ----- 9.451268005811858
----- Test RMSE for EMA----- 4.930947218202654 ----- Test_MSE_LSTM for EMA----- 24.31424046870049 ----- Test_MAE_LSTM for EMA----- 4.072073008160127
----- Train RMSE for WMA ----- 10.97436711794096 ----- Train_MSE_LSTM for WMA ----- 120.43673363934376 ----- Train MAE LSTM for WMA ----- 9.925815236695067
----- Test RMSE for WMA----- 8.255404404296101 ----- Test_MSE_LSTM for WMA----- 68.15170187847146 ----- Test_MAE_LSTM for WMA----- 6.806281257852447
----- Train RMSE for DEMA ----- 12.066299877592092 ----- Train_MSE_LSTM for DEMA ----- 145.59559273597895 ----- Train MAE LSTM for DEMA ----- 10.907355915123395
----- Test RMSE for DEMA----- 6.627380888958364 ----- Test_MSE_LSTM for DEMA----- 43.922177447330554 ----- Test_MAE_LSTM for DEMA----- 5.414540694927783
----- Train RMSE for KAMA ----- 10.869146791217638 ----- Train_MSE_LSTM for KAMA ----- 118.13835196903668 ----- Train MAE LSTM for KAMA ----- 9.759291692937742
----- Test RMSE for KAMA----- 4.8986153334270215 ----- Test_MSE_LSTM for KAMA----- 23.99643218488633 ----- Test_MAE_LSTM for KAMA----- 3.8674202618000764
----- Train RMSE for MIDPOINT ----- 9.691011765630286 ----- Train_MSE_LSTM for MIDPOINT ----- 93.91570904158462 ----- Train MAE LSTM for MIDPOINT ----- 8.650189046672622
----- Test RMSE for MIDPOINT----- 5.359026541555295 ----- Test_MSE_LSTM for MIDPOINT----- 28.7191654730941 ----- Test_MAE_LSTM for MIDPOINT----- 4.42030651732032
----- Train RMSE for T3 ----- 12.129650565307267 ----- Train_MSE_LSTM for T3 ----- 147.12842283645892 ----- Train MAE LSTM for T3 ----- 10.936634904781638
----- Test RMSE for T3----- 7.553602318843705 ----- Test_MSE_LSTM for T3----- 57.05690799124099 ----- Test_MAE_LSTM for T3----- 6.064989593585796
----- Train RMSE for TEMA ----- 7.376932734913034 ----- Train_MSE_LSTM for TEMA ----- 54.419136575431494 ----- Train MAE LSTM for TEMA ----- 5.050424731629302
----- Test RMSE for TEMA----- 4.578721256009806 ----- Test_MSE_LSTM for TEMA----- 20.964688340236016 ----- Test_MAE_LSTM for TEMA----- 3.7212897315589664
Univariate Arima Multistep MutiVariate LSTM Hybrid Model Experiment 4¶
From the above experiments it is evident that with Higher moving averages the loss plots show unreoresented data and underfitting, hence keeping only the MA's that have smaller periods like T3 OR TRIMA. Going forward EMA, WMA & DEMA will be ignored.
In [ ]:
def get_lstm(data,original_data, train_len, test_len,img_file,ma ,lstm_len=3):
# prepare train and test data
X_value = pd.DataFrame(data.iloc[:, :])
y_value = pd.DataFrame(data.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
# Get data and check shape
X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X)
y_train, y_test, = split_train_test(y)
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
det = 20
input_dim = X_train.shape[1]#3
feature_size = X_train.shape[2]#24
output_dim = y_train.shape[1]#1
# # Option 1
# # Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(256, activation='relu', kernel_initializer='he_normal', input_shape=(input_dim, feature_size)))
# model.add(Dense(units=64,activation='relu'))
# model.add(Dropout(0.5))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(learning_rate = 0.001), loss='mse')
# ## Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# # option 2
# model = Sequential()
# model.add(Bidirectional(LSTM(units= 128), input_shape=(input_dim, feature_size)))
# model.add(Dense(64))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(lr = 0.001), loss='mean_squared_error', metrics=['accuracy'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# # Option 3
# # define custom activation
# #
# class Double_Tanh(Activation):
# def __init__(self, activation, **kwargs):
# super(Double_Tanh, self).__init__(activation, **kwargs)
# self.__name__ = 'double_tanh'
# def double_tanh(x):
# return (K.tanh(x) * 2)
# get_custom_objects().update({'double_tanh':Double_Tanh(double_tanh)})
# # Model Generation
# model = Sequential()
# #check https://machinelearningmastery.com/use-weight-regularization-lstm-networks-time-series-forecasting/
# model.add(LSTM(25, input_shape=(input_dim, feature_size), dropout=0.2, kernel_regularizer=l1_l2(0.00,0.00), bias_regularizer=l1_l2(0.00,0.00)))
# model.add(Dense(1))
# model.add(Activation(double_tanh))
# model.compile(loss='mean_squared_error', optimizer='adam', metrics=['mse', 'mae'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# Option 4
# Set up & fit LSTM RNN
model = Sequential()
model.add(LSTM(units=lstm_len, return_sequences=True, input_shape=(input_dim, feature_size)))
model.add(LSTM(units=int(lstm_len/2)))
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='mean_squared_error', optimizer='adam')
# Common code
callbacks = [
EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
ModelCheckpoint('LSTM4.h5', verbose=1, save_best_only=True, save_weights_only=True)]
fname1 = img_file+'.png'
tensorflow.keras.utils.plot_model(
model, to_file=fname1, show_shapes=True, show_dtype=False,
show_layer_names=True, expand_nested=False, dpi=96,
layer_range=None, show_layer_activations=False
)
history = model.fit(X_train, y_train, epochs=500, batch_size=int( optimized_period[ma]), verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# plot loss
fname2 = img_file+'-'+ma
plt.title(img_file+'-'+ma+' Loss')
plt.xlabel("Epochs")
plt.ylabel("Loss")
pyplot.plot(history.history['loss'], label='train')
pyplot.plot(history.history['val_loss'], label='validation')
pyplot.legend()
pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# Generate predictions
predictiontr = model.predict(X_train, verbose=0)
predictiontr = (y_scaler.inverse_transform(predictiontr)-det).tolist()
outputtr = []
for i in range(len(predictiontr)):
outputtr.extend(predictiontr[i])
predictiontr = outputtr
# Generate error data
## replace with yc , xtest generated by new multistep method
mse_tr = mean_squared_error(y_train, predictiontr)
rmse_tr = mse_tr ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictiontr))
mae_tr = mean_absolute_error(y_train, pd.Series(predictiontr))
# Original_tr = pd.Series(yc_train)
Original_tr = y_scaler.inverse_transform(y_train).flatten().tolist()
predictionte = model.predict(X_test, verbose=0)
predictionte =( y_scaler.inverse_transform(predictionte)-det).tolist()
outputte = []
for i in range(len(predictionte)):
outputte.extend(predictionte[i])
predictionte = outputte
# Generate error data
mse_te = mean_squared_error(y_test, predictionte)
rmse_te = mse_te ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictionte))
mae_te = mean_absolute_error(y_test, pd.Series(predictionte))
# Original_te = pd.Series(yc_test)
Original_te = y_scaler.inverse_transform(y_test).flatten().tolist()
return Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,Original_te,predictionte, mse_te,rmse_te,mae_te
In [ ]:
if __name__ == '__main__':
start_time = timeit.default_timer()
simulation4 = {}
imgfile = 'Experiment4'
for ma in optimized_period:
print(ma)
print(functions[ma])
print ( int( optimized_period[ma]))
# if ma == 'SMA':
low_vol = df.apply(lambda c: functions[ma](c, timeperiod = int( optimized_period[ma])))
low_vol = low_vol.fillna(0)
low_vol_data = df['close']
high_vol = pd.DataFrame()
df2 = df.copy()
for i in df2.columns:
if i in low_vol.columns:
high_vol[i] = df2[i].subtract(low_vol[i], fill_value=0)
high_vol_data = df['close']
## *****************************************************
# Generate ARIMA and LSTM predictions
print('\nWorking on ' + ma + ' predictions')
try:
print('parameters used : ', train_len, test_len)
low_vol_Original, low_vol_prediction, low_vol_mse, low_vol_rmse,low_vol_mae = get_arima(low_vol,low_vol_data, train_len, test_len)
except:
print('ARIMA error, skipping to next MA type')
continue
Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,high_vol_Original, high_vol_prediction, high_vol_mse, high_vol_rmse,high_vol_mae, = get_lstm(high_vol,high_vol_data, train_len, test_len,imgfile,ma)
final_prediction_tr = df['close'].head(train_len).values + pd.Series(predictiontr) # ignoring first 3 steps
mse_ftr = mean_squared_error(df['close'].head(train_len).values,final_prediction_tr.values)
rmse_ftr = mse_ftr ** 0.5
mape_ftr = mean_absolute_percentage_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
mae_ftr = mean_absolute_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
final_prediction = pd.Series(low_vol_prediction[3:]) + pd.Series(high_vol_prediction)
mse = mean_squared_error(df['close'].tail(test_len).values,final_prediction.values)
rmse = mse ** 0.5
mape = mean_absolute_percentage_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
mae = mean_absolute_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
# Generate prediction accuracy
actual = df['close'].tail(test_len).values
result_1 = []
result_2 = []
for i in range(1, len(final_prediction)):
# Compare prediction to previous close price
if final_prediction[i] > actual[i-1] and actual[i] > actual[i-1]:
result_1.append(1)
elif final_prediction[i] < actual[i-1] and actual[i] < actual[i-1]:
result_1.append(1)
else:
result_1.append(0)
# Compare prediction to previous prediction
if final_prediction[i] > final_prediction[i-1] and actual[i] > actual[i-1]:
result_2.append(1)
elif final_prediction[i] < final_prediction[i-1] and actual[i] < actual[i-1]:
result_2.append(1)
else:
result_2.append(0)
accuracy_1 = np.mean(result_1)
accuracy_2 = np.mean(result_2)
simulation4[ma] = {'low_vol': {'original':list(low_vol_Original), 'prediction': list(low_vol_prediction) , 'mse': low_vol_mse,
'rmse': low_vol_rmse, 'mae' : low_vol_mae},
'high_vol': {'original':list(high_vol_Original),'prediction': list(high_vol_prediction), 'mse': high_vol_mse,
'rmse': high_vol_rmse, 'mae' : high_vol_mae},
'final_tr': {'original':df['close'].head(train_len).tolist(),'prediction': final_prediction_tr.values.tolist(), 'mse': mse_ftr,
'rmse': rmse_ftr, 'mae' : mae_ftr},
'final': {'original': df['close'].tail(test_len).tolist(), 'prediction': final_prediction.values.tolist(), 'mse': mse,
'rmse': rmse, 'mae': mae },
'accuracy': {'prediction vs close': accuracy_1, 'prediction vs prediction': accuracy_2}}
# save simulation data here as checkpoint
with open('simulation4_data.json', 'w') as fp:
json.dump(simulation4, fp)
for ma in simulation4.keys():
print('\n' + ma)
print('Prediction vs Close:\t\t' + str(round(100*simulation4[ma]['accuracy']['prediction vs close'], 2))
+ '% Accuracy')
print('Prediction vs Prediction:\t' + str(round(100*simulation4[ma]['accuracy']['prediction vs prediction'], 2))
+ '% Accuracy')
print('MSE:\t', simulation4[ma]['final']['mse'],
'\nRMSE:\t', simulation4[ma]['final']['rmse'],
'\nMAPE:\t', simulation4[ma]['final']['mae'])#,
# '\nMAPE:\t', simulation[ma]['final']['mape'])
# else:
# break
elapsed = timeit.default_timer() - start_time
print('Runtime: mins:',elapsed/60)
SMA
SMA([input_arrays], [timeperiod=30])
Simple Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
17
Working on SMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.37 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4157.020, Time=0.02 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3687.148, Time=0.02 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.12 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3458.651, Time=0.04 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3322.133, Time=0.06 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=0.39 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.44 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3324.133, Time=0.12 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 1.570 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1657.067
Date: Sun, 12 Dec 2021 AIC 3322.133
Time: 13:12:09 BIC 3340.897
Sample: 0 HQIC 3329.339
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1966 0.003 -387.226 0.000 -1.203 -1.191
ar.L2 -0.8952 0.006 -138.692 0.000 -0.908 -0.883
ar.L3 -0.3968 0.006 -68.284 0.000 -0.408 -0.385
sigma2 3.5858 0.017 214.535 0.000 3.553 3.619
===================================================================================
Ljung-Box (L1) (Q): 14.47 Jarque-Bera (JB): 2428881.42
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 271.99
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.05330, saving model to LSTM4.h5
48/48 - 6s - loss: 1.4217 - val_loss: 0.0533 - lr: 0.0010 - 6s/epoch - 129ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.05330
48/48 - 0s - loss: 1.3808 - val_loss: 0.0552 - lr: 0.0010 - 223ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.05330
48/48 - 0s - loss: 1.2868 - val_loss: 0.0567 - lr: 0.0010 - 218ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.05330
48/48 - 0s - loss: 1.1463 - val_loss: 0.0611 - lr: 0.0010 - 222ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.05330
48/48 - 0s - loss: 1.0599 - val_loss: 0.0659 - lr: 0.0010 - 220ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.05330
48/48 - 0s - loss: 1.0010 - val_loss: 0.0709 - lr: 0.0010 - 227ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9705 - val_loss: 0.0714 - lr: 1.0000e-04 - 223ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9661 - val_loss: 0.0719 - lr: 1.0000e-04 - 215ms/epoch - 4ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9619 - val_loss: 0.0725 - lr: 1.0000e-04 - 226ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9577 - val_loss: 0.0730 - lr: 1.0000e-04 - 227ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9537 - val_loss: 0.0736 - lr: 1.0000e-04 - 222ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9511 - val_loss: 0.0737 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9507 - val_loss: 0.0737 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9503 - val_loss: 0.0738 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9499 - val_loss: 0.0738 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9495 - val_loss: 0.0739 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9491 - val_loss: 0.0740 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9487 - val_loss: 0.0740 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9483 - val_loss: 0.0741 - lr: 1.0000e-05 - 230ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9479 - val_loss: 0.0742 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9475 - val_loss: 0.0742 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9471 - val_loss: 0.0743 - lr: 1.0000e-05 - 216ms/epoch - 4ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9466 - val_loss: 0.0744 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9462 - val_loss: 0.0744 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9458 - val_loss: 0.0745 - lr: 1.0000e-05 - 211ms/epoch - 4ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9454 - val_loss: 0.0746 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9450 - val_loss: 0.0747 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9446 - val_loss: 0.0747 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9442 - val_loss: 0.0748 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9438 - val_loss: 0.0749 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9433 - val_loss: 0.0750 - lr: 1.0000e-05 - 212ms/epoch - 4ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9429 - val_loss: 0.0750 - lr: 1.0000e-05 - 213ms/epoch - 4ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9425 - val_loss: 0.0751 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9421 - val_loss: 0.0752 - lr: 1.0000e-05 - 215ms/epoch - 4ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9417 - val_loss: 0.0753 - lr: 1.0000e-05 - 207ms/epoch - 4ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9413 - val_loss: 0.0753 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9409 - val_loss: 0.0754 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9405 - val_loss: 0.0755 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9400 - val_loss: 0.0756 - lr: 1.0000e-05 - 208ms/epoch - 4ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9396 - val_loss: 0.0756 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9392 - val_loss: 0.0757 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9388 - val_loss: 0.0758 - lr: 1.0000e-05 - 216ms/epoch - 4ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9384 - val_loss: 0.0759 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9380 - val_loss: 0.0760 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9376 - val_loss: 0.0761 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9371 - val_loss: 0.0761 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9367 - val_loss: 0.0762 - lr: 1.0000e-05 - 227ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9363 - val_loss: 0.0763 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9359 - val_loss: 0.0764 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9355 - val_loss: 0.0765 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.05330
48/48 - 0s - loss: 0.9351 - val_loss: 0.0766 - lr: 1.0000e-05 - 215ms/epoch - 4ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 50.37% Accuracy
MSE: 23.31830336349202
RMSE: 4.828902915103183
MAPE: 3.806885992834059
EMA
EMA([input_arrays], [timeperiod=30])
Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
51
Working on EMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.30 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4231.556, Time=0.02 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3761.238, Time=0.02 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.15 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3532.227, Time=0.04 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3394.496, Time=0.09 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=0.81 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.41 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3396.496, Time=0.21 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 2.055 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1693.248
Date: Sun, 12 Dec 2021 AIC 3394.496
Time: 13:13:27 BIC 3413.260
Sample: 0 HQIC 3401.702
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1982 0.003 -389.569 0.000 -1.204 -1.192
ar.L2 -0.8976 0.006 -139.811 0.000 -0.910 -0.885
ar.L3 -0.3984 0.006 -68.662 0.000 -0.410 -0.387
sigma2 3.9230 0.018 215.372 0.000 3.887 3.959
===================================================================================
Ljung-Box (L1) (Q): 14.54 Jarque-Bera (JB): 2462173.05
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 273.82
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.05737, saving model to LSTM4.h5
16/16 - 3s - loss: 1.5587 - val_loss: 0.0574 - lr: 0.0010 - 3s/epoch - 208ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.5128 - val_loss: 0.0594 - lr: 0.0010 - 88ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.4708 - val_loss: 0.0613 - lr: 0.0010 - 87ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.4320 - val_loss: 0.0632 - lr: 0.0010 - 93ms/epoch - 6ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3971 - val_loss: 0.0651 - lr: 0.0010 - 100ms/epoch - 6ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3669 - val_loss: 0.0669 - lr: 0.0010 - 91ms/epoch - 6ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3490 - val_loss: 0.0671 - lr: 1.0000e-04 - 88ms/epoch - 6ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3465 - val_loss: 0.0672 - lr: 1.0000e-04 - 84ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3440 - val_loss: 0.0674 - lr: 1.0000e-04 - 99ms/epoch - 6ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3415 - val_loss: 0.0676 - lr: 1.0000e-04 - 89ms/epoch - 6ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3391 - val_loss: 0.0678 - lr: 1.0000e-04 - 95ms/epoch - 6ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3375 - val_loss: 0.0678 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3372 - val_loss: 0.0678 - lr: 1.0000e-05 - 89ms/epoch - 6ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3370 - val_loss: 0.0679 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3367 - val_loss: 0.0679 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3365 - val_loss: 0.0679 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3363 - val_loss: 0.0679 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3360 - val_loss: 0.0679 - lr: 1.0000e-05 - 101ms/epoch - 6ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3358 - val_loss: 0.0680 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3355 - val_loss: 0.0680 - lr: 1.0000e-05 - 92ms/epoch - 6ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3353 - val_loss: 0.0680 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3350 - val_loss: 0.0680 - lr: 1.0000e-05 - 88ms/epoch - 6ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3348 - val_loss: 0.0680 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3345 - val_loss: 0.0681 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3343 - val_loss: 0.0681 - lr: 1.0000e-05 - 88ms/epoch - 6ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3340 - val_loss: 0.0681 - lr: 1.0000e-05 - 88ms/epoch - 6ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3338 - val_loss: 0.0681 - lr: 1.0000e-05 - 89ms/epoch - 6ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3335 - val_loss: 0.0681 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3333 - val_loss: 0.0682 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3330 - val_loss: 0.0682 - lr: 1.0000e-05 - 81ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3328 - val_loss: 0.0682 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3325 - val_loss: 0.0682 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3323 - val_loss: 0.0683 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3320 - val_loss: 0.0683 - lr: 1.0000e-05 - 89ms/epoch - 6ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3318 - val_loss: 0.0683 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3315 - val_loss: 0.0683 - lr: 1.0000e-05 - 89ms/epoch - 6ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3313 - val_loss: 0.0684 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3310 - val_loss: 0.0684 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3308 - val_loss: 0.0684 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3305 - val_loss: 0.0684 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3303 - val_loss: 0.0684 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3300 - val_loss: 0.0685 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3298 - val_loss: 0.0685 - lr: 1.0000e-05 - 81ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3295 - val_loss: 0.0685 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3293 - val_loss: 0.0685 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3290 - val_loss: 0.0686 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3288 - val_loss: 0.0686 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3285 - val_loss: 0.0686 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3283 - val_loss: 0.0686 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3280 - val_loss: 0.0686 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.05737
16/16 - 0s - loss: 1.3278 - val_loss: 0.0687 - lr: 1.0000e-05 - 93ms/epoch - 6ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 50.37% Accuracy
MSE: 23.31830336349202
RMSE: 4.828902915103183
MAPE: 3.806885992834059
EMA
Prediction vs Close: 55.97% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 31.4391560756623
RMSE: 5.607063052584865
MAPE: 4.398444723456604
WMA
WMA([input_arrays], [timeperiod=30])
Weighted Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
49
Working on WMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.28 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4264.089, Time=0.02 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3793.930, Time=0.02 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.13 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3564.923, Time=0.06 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3427.258, Time=0.08 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.19 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.31 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3429.258, Time=0.17 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 2.256 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1709.629
Date: Sun, 12 Dec 2021 AIC 3427.258
Time: 13:14:36 BIC 3446.021
Sample: 0 HQIC 3434.464
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1981 0.003 -389.386 0.000 -1.204 -1.192
ar.L2 -0.8974 0.006 -139.699 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.737 0.000 -0.410 -0.387
sigma2 4.0860 0.019 215.311 0.000 4.049 4.123
===================================================================================
Ljung-Box (L1) (Q): 14.57 Jarque-Bera (JB): 2460901.70
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 273.75
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04825, saving model to LSTM4.h5
17/17 - 3s - loss: 1.4220 - val_loss: 0.0483 - lr: 0.0010 - 3s/epoch - 199ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.4044 - val_loss: 0.0489 - lr: 0.0010 - 93ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.3853 - val_loss: 0.0496 - lr: 0.0010 - 95ms/epoch - 6ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.3646 - val_loss: 0.0504 - lr: 0.0010 - 87ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.3430 - val_loss: 0.0512 - lr: 0.0010 - 93ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.3211 - val_loss: 0.0521 - lr: 0.0010 - 90ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.3068 - val_loss: 0.0522 - lr: 1.0000e-04 - 95ms/epoch - 6ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.3047 - val_loss: 0.0523 - lr: 1.0000e-04 - 94ms/epoch - 6ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.3026 - val_loss: 0.0524 - lr: 1.0000e-04 - 92ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.3005 - val_loss: 0.0525 - lr: 1.0000e-04 - 88ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2985 - val_loss: 0.0526 - lr: 1.0000e-04 - 90ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2972 - val_loss: 0.0526 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2970 - val_loss: 0.0526 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2968 - val_loss: 0.0526 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2966 - val_loss: 0.0526 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2964 - val_loss: 0.0526 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2962 - val_loss: 0.0526 - lr: 1.0000e-05 - 88ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2960 - val_loss: 0.0526 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2958 - val_loss: 0.0526 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2956 - val_loss: 0.0527 - lr: 1.0000e-05 - 91ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2954 - val_loss: 0.0527 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2952 - val_loss: 0.0527 - lr: 1.0000e-05 - 90ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2950 - val_loss: 0.0527 - lr: 1.0000e-05 - 90ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2948 - val_loss: 0.0527 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2946 - val_loss: 0.0527 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2944 - val_loss: 0.0527 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2942 - val_loss: 0.0527 - lr: 1.0000e-05 - 98ms/epoch - 6ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2940 - val_loss: 0.0527 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2938 - val_loss: 0.0528 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2936 - val_loss: 0.0528 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2934 - val_loss: 0.0528 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2932 - val_loss: 0.0528 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2931 - val_loss: 0.0528 - lr: 1.0000e-05 - 98ms/epoch - 6ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2929 - val_loss: 0.0528 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2927 - val_loss: 0.0528 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2925 - val_loss: 0.0528 - lr: 1.0000e-05 - 88ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2923 - val_loss: 0.0528 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2921 - val_loss: 0.0528 - lr: 1.0000e-05 - 100ms/epoch - 6ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2919 - val_loss: 0.0529 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2917 - val_loss: 0.0529 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2915 - val_loss: 0.0529 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2913 - val_loss: 0.0529 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2911 - val_loss: 0.0529 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2909 - val_loss: 0.0529 - lr: 1.0000e-05 - 105ms/epoch - 6ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2907 - val_loss: 0.0529 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2905 - val_loss: 0.0529 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2903 - val_loss: 0.0529 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2901 - val_loss: 0.0530 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2899 - val_loss: 0.0530 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2898 - val_loss: 0.0530 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.04825
17/17 - 0s - loss: 1.2896 - val_loss: 0.0530 - lr: 1.0000e-05 - 98ms/epoch - 6ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 50.37% Accuracy
MSE: 23.31830336349202
RMSE: 4.828902915103183
MAPE: 3.806885992834059
EMA
Prediction vs Close: 55.97% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 31.4391560756623
RMSE: 5.607063052584865
MAPE: 4.398444723456604
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 48.85272948439791
RMSE: 6.989472761546318
MAPE: 5.616901258925532
DEMA
DEMA([input_arrays], [timeperiod=30])
Double Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
89
Working on DEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.28 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4436.126, Time=0.02 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3965.317, Time=0.02 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.22 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3736.589, Time=0.04 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3598.951, Time=0.05 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=0.53 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.52 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3600.951, Time=0.12 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 1.791 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1795.475
Date: Sun, 12 Dec 2021 AIC 3598.951
Time: 13:15:44 BIC 3617.714
Sample: 0 HQIC 3606.157
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1983 0.003 -389.581 0.000 -1.204 -1.192
ar.L2 -0.8973 0.006 -139.732 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.649 0.000 -0.410 -0.387
sigma2 5.0573 0.023 215.292 0.000 5.011 5.103
===================================================================================
Ljung-Box (L1) (Q): 14.41 Jarque-Bera (JB): 2460553.80
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.89
Prob(H) (two-sided): 0.00 Kurtosis: 273.74
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04932, saving model to LSTM4.h5
10/10 - 4s - loss: 1.4315 - val_loss: 0.0493 - lr: 0.0010 - 4s/epoch - 380ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.4228 - val_loss: 0.0499 - lr: 0.0010 - 69ms/epoch - 7ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.4140 - val_loss: 0.0505 - lr: 0.0010 - 62ms/epoch - 6ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.4050 - val_loss: 0.0511 - lr: 0.0010 - 66ms/epoch - 7ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3959 - val_loss: 0.0517 - lr: 0.0010 - 59ms/epoch - 6ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3866 - val_loss: 0.0523 - lr: 0.0010 - 66ms/epoch - 7ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3800 - val_loss: 0.0524 - lr: 1.0000e-04 - 65ms/epoch - 7ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3790 - val_loss: 0.0524 - lr: 1.0000e-04 - 60ms/epoch - 6ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3781 - val_loss: 0.0525 - lr: 1.0000e-04 - 63ms/epoch - 6ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3771 - val_loss: 0.0525 - lr: 1.0000e-04 - 70ms/epoch - 7ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3762 - val_loss: 0.0526 - lr: 1.0000e-04 - 64ms/epoch - 6ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3755 - val_loss: 0.0526 - lr: 1.0000e-05 - 65ms/epoch - 6ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3754 - val_loss: 0.0526 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3753 - val_loss: 0.0526 - lr: 1.0000e-05 - 65ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3752 - val_loss: 0.0526 - lr: 1.0000e-05 - 65ms/epoch - 7ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3751 - val_loss: 0.0526 - lr: 1.0000e-05 - 69ms/epoch - 7ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3750 - val_loss: 0.0526 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3749 - val_loss: 0.0526 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3749 - val_loss: 0.0526 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3748 - val_loss: 0.0526 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3747 - val_loss: 0.0526 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3746 - val_loss: 0.0526 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3745 - val_loss: 0.0526 - lr: 1.0000e-05 - 65ms/epoch - 7ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3744 - val_loss: 0.0526 - lr: 1.0000e-05 - 67ms/epoch - 7ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3743 - val_loss: 0.0527 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3742 - val_loss: 0.0527 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3741 - val_loss: 0.0527 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3740 - val_loss: 0.0527 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3739 - val_loss: 0.0527 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3738 - val_loss: 0.0527 - lr: 1.0000e-05 - 65ms/epoch - 7ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3737 - val_loss: 0.0527 - lr: 1.0000e-05 - 72ms/epoch - 7ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3736 - val_loss: 0.0527 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3735 - val_loss: 0.0527 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3734 - val_loss: 0.0527 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3734 - val_loss: 0.0527 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3733 - val_loss: 0.0527 - lr: 1.0000e-05 - 62ms/epoch - 6ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3732 - val_loss: 0.0527 - lr: 1.0000e-05 - 71ms/epoch - 7ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3731 - val_loss: 0.0527 - lr: 1.0000e-05 - 64ms/epoch - 6ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3730 - val_loss: 0.0527 - lr: 1.0000e-05 - 67ms/epoch - 7ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3729 - val_loss: 0.0527 - lr: 1.0000e-05 - 62ms/epoch - 6ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3728 - val_loss: 0.0527 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3727 - val_loss: 0.0527 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3726 - val_loss: 0.0527 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3725 - val_loss: 0.0527 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3724 - val_loss: 0.0528 - lr: 1.0000e-05 - 57ms/epoch - 6ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3723 - val_loss: 0.0528 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3722 - val_loss: 0.0528 - lr: 1.0000e-05 - 64ms/epoch - 6ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3721 - val_loss: 0.0528 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3720 - val_loss: 0.0528 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3719 - val_loss: 0.0528 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.04932
10/10 - 0s - loss: 1.3718 - val_loss: 0.0528 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 50.37% Accuracy
MSE: 23.31830336349202
RMSE: 4.828902915103183
MAPE: 3.806885992834059
EMA
Prediction vs Close: 55.97% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 31.4391560756623
RMSE: 5.607063052584865
MAPE: 4.398444723456604
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 48.85272948439791
RMSE: 6.989472761546318
MAPE: 5.616901258925532
DEMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 143.4471215002686
RMSE: 11.976941241413376
MAPE: 10.686872819228396
KAMA
KAMA([input_arrays], [timeperiod=30])
Kaufman Adaptive Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
18
Working on KAMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.25 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4190.464, Time=0.02 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3724.371, Time=0.02 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.15 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3494.154, Time=0.04 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3357.435, Time=0.09 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=0.71 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.42 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3359.435, Time=0.12 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 1.825 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1674.717
Date: Sun, 12 Dec 2021 AIC 3357.435
Time: 13:16:44 BIC 3376.198
Sample: 0 HQIC 3364.641
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1955 0.003 -381.246 0.000 -1.202 -1.189
ar.L2 -0.8964 0.007 -135.835 0.000 -0.909 -0.883
ar.L3 -0.3971 0.006 -67.229 0.000 -0.409 -0.385
sigma2 3.7466 0.018 211.623 0.000 3.712 3.781
===================================================================================
Ljung-Box (L1) (Q): 14.20 Jarque-Bera (JB): 2338363.32
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.01 Skew: 3.76
Prob(H) (two-sided): 0.00 Kurtosis: 266.93
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04951, saving model to LSTM4.h5
45/45 - 4s - loss: 1.3909 - val_loss: 0.0495 - lr: 0.0010 - 4s/epoch - 84ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.04951
45/45 - 0s - loss: 1.2737 - val_loss: 0.0516 - lr: 0.0010 - 226ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04951
45/45 - 0s - loss: 1.1363 - val_loss: 0.0550 - lr: 0.0010 - 231ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04951
45/45 - 0s - loss: 1.0490 - val_loss: 0.0585 - lr: 0.0010 - 209ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9940 - val_loss: 0.0621 - lr: 0.0010 - 205ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9525 - val_loss: 0.0658 - lr: 0.0010 - 222ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9304 - val_loss: 0.0662 - lr: 1.0000e-04 - 228ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9271 - val_loss: 0.0666 - lr: 1.0000e-04 - 220ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9238 - val_loss: 0.0670 - lr: 1.0000e-04 - 220ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9205 - val_loss: 0.0674 - lr: 1.0000e-04 - 205ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9172 - val_loss: 0.0678 - lr: 1.0000e-04 - 230ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9152 - val_loss: 0.0678 - lr: 1.0000e-05 - 228ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9149 - val_loss: 0.0679 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9146 - val_loss: 0.0679 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9142 - val_loss: 0.0680 - lr: 1.0000e-05 - 203ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9139 - val_loss: 0.0680 - lr: 1.0000e-05 - 238ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9136 - val_loss: 0.0681 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9132 - val_loss: 0.0681 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9129 - val_loss: 0.0682 - lr: 1.0000e-05 - 214ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9126 - val_loss: 0.0682 - lr: 1.0000e-05 - 230ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9122 - val_loss: 0.0683 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9119 - val_loss: 0.0683 - lr: 1.0000e-05 - 203ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9115 - val_loss: 0.0684 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9112 - val_loss: 0.0684 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9108 - val_loss: 0.0685 - lr: 1.0000e-05 - 235ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9105 - val_loss: 0.0685 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9101 - val_loss: 0.0686 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9098 - val_loss: 0.0686 - lr: 1.0000e-05 - 198ms/epoch - 4ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9094 - val_loss: 0.0687 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9091 - val_loss: 0.0687 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9087 - val_loss: 0.0688 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9084 - val_loss: 0.0689 - lr: 1.0000e-05 - 204ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9080 - val_loss: 0.0689 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9077 - val_loss: 0.0690 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9073 - val_loss: 0.0690 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9070 - val_loss: 0.0691 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9066 - val_loss: 0.0692 - lr: 1.0000e-05 - 204ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9063 - val_loss: 0.0692 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9059 - val_loss: 0.0693 - lr: 1.0000e-05 - 236ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9056 - val_loss: 0.0693 - lr: 1.0000e-05 - 200ms/epoch - 4ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9052 - val_loss: 0.0694 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9048 - val_loss: 0.0695 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9045 - val_loss: 0.0695 - lr: 1.0000e-05 - 208ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9041 - val_loss: 0.0696 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9038 - val_loss: 0.0696 - lr: 1.0000e-05 - 212ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9034 - val_loss: 0.0697 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9031 - val_loss: 0.0698 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9027 - val_loss: 0.0698 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9024 - val_loss: 0.0699 - lr: 1.0000e-05 - 212ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9020 - val_loss: 0.0700 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.04951
45/45 - 0s - loss: 0.9017 - val_loss: 0.0700 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 50.37% Accuracy
MSE: 23.31830336349202
RMSE: 4.828902915103183
MAPE: 3.806885992834059
EMA
Prediction vs Close: 55.97% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 31.4391560756623
RMSE: 5.607063052584865
MAPE: 4.398444723456604
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 48.85272948439791
RMSE: 6.989472761546318
MAPE: 5.616901258925532
DEMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 143.4471215002686
RMSE: 11.976941241413376
MAPE: 10.686872819228396
KAMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 23.251670970583447
RMSE: 4.821998648961181
MAPE: 3.833042253232743
MIDPOINT
MIDPOINT([input_arrays], [timeperiod=14])
MidPoint over period (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 14
Outputs:
real
14
Working on MIDPOINT predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.26 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4212.289, Time=0.02 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3747.746, Time=0.03 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.14 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3523.401, Time=0.04 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3387.759, Time=0.05 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=0.69 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.50 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3389.758, Time=0.21 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 1.941 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1689.879
Date: Sun, 12 Dec 2021 AIC 3387.759
Time: 13:18:02 BIC 3406.522
Sample: 0 HQIC 3394.964
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1878 0.003 -345.315 0.000 -1.195 -1.181
ar.L2 -0.8876 0.007 -121.809 0.000 -0.902 -0.873
ar.L3 -0.3957 0.007 -60.127 0.000 -0.409 -0.383
sigma2 3.8904 0.020 193.404 0.000 3.851 3.930
===================================================================================
Ljung-Box (L1) (Q): 13.21 Jarque-Bera (JB): 1659080.01
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.08 Skew: 3.28
Prob(H) (two-sided): 0.00 Kurtosis: 225.31
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.06096, saving model to LSTM4.h5
58/58 - 4s - loss: 1.4939 - val_loss: 0.0610 - lr: 0.0010 - 4s/epoch - 62ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.06096
58/58 - 0s - loss: 1.2750 - val_loss: 0.0712 - lr: 0.0010 - 298ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.06096
58/58 - 0s - loss: 1.0964 - val_loss: 0.0798 - lr: 0.0010 - 271ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.9939 - val_loss: 0.0866 - lr: 0.0010 - 272ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.9161 - val_loss: 0.0916 - lr: 0.0010 - 274ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.8398 - val_loss: 0.0966 - lr: 0.0010 - 269ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7994 - val_loss: 0.0971 - lr: 1.0000e-04 - 258ms/epoch - 4ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7951 - val_loss: 0.0976 - lr: 1.0000e-04 - 272ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7910 - val_loss: 0.0982 - lr: 1.0000e-04 - 275ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7871 - val_loss: 0.0988 - lr: 1.0000e-04 - 279ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7834 - val_loss: 0.0994 - lr: 1.0000e-04 - 259ms/epoch - 4ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7811 - val_loss: 0.0995 - lr: 1.0000e-05 - 273ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7807 - val_loss: 0.0995 - lr: 1.0000e-05 - 286ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7804 - val_loss: 0.0996 - lr: 1.0000e-05 - 274ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7800 - val_loss: 0.0997 - lr: 1.0000e-05 - 269ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7796 - val_loss: 0.0997 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7793 - val_loss: 0.0998 - lr: 1.0000e-05 - 285ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7789 - val_loss: 0.0999 - lr: 1.0000e-05 - 266ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7785 - val_loss: 0.0999 - lr: 1.0000e-05 - 256ms/epoch - 4ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7781 - val_loss: 0.1000 - lr: 1.0000e-05 - 301ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7777 - val_loss: 0.1001 - lr: 1.0000e-05 - 283ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7773 - val_loss: 0.1002 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7770 - val_loss: 0.1002 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7766 - val_loss: 0.1003 - lr: 1.0000e-05 - 285ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7762 - val_loss: 0.1004 - lr: 1.0000e-05 - 263ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7758 - val_loss: 0.1005 - lr: 1.0000e-05 - 260ms/epoch - 4ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7754 - val_loss: 0.1006 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7750 - val_loss: 0.1006 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7746 - val_loss: 0.1007 - lr: 1.0000e-05 - 260ms/epoch - 4ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7742 - val_loss: 0.1008 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7738 - val_loss: 0.1009 - lr: 1.0000e-05 - 291ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7734 - val_loss: 0.1010 - lr: 1.0000e-05 - 267ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7730 - val_loss: 0.1010 - lr: 1.0000e-05 - 278ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7726 - val_loss: 0.1011 - lr: 1.0000e-05 - 267ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7722 - val_loss: 0.1012 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7718 - val_loss: 0.1013 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7714 - val_loss: 0.1014 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7710 - val_loss: 0.1014 - lr: 1.0000e-05 - 258ms/epoch - 4ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7706 - val_loss: 0.1015 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7702 - val_loss: 0.1016 - lr: 1.0000e-05 - 256ms/epoch - 4ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7698 - val_loss: 0.1017 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7695 - val_loss: 0.1018 - lr: 1.0000e-05 - 275ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7691 - val_loss: 0.1019 - lr: 1.0000e-05 - 288ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7687 - val_loss: 0.1020 - lr: 1.0000e-05 - 269ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7683 - val_loss: 0.1020 - lr: 1.0000e-05 - 274ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7679 - val_loss: 0.1021 - lr: 1.0000e-05 - 290ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7675 - val_loss: 0.1022 - lr: 1.0000e-05 - 275ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7671 - val_loss: 0.1023 - lr: 1.0000e-05 - 269ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7667 - val_loss: 0.1024 - lr: 1.0000e-05 - 275ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7663 - val_loss: 0.1025 - lr: 1.0000e-05 - 286ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.06096
58/58 - 0s - loss: 0.7659 - val_loss: 0.1026 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 50.37% Accuracy
MSE: 23.31830336349202
RMSE: 4.828902915103183
MAPE: 3.806885992834059
EMA
Prediction vs Close: 55.97% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 31.4391560756623
RMSE: 5.607063052584865
MAPE: 4.398444723456604
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 48.85272948439791
RMSE: 6.989472761546318
MAPE: 5.616901258925532
DEMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 143.4471215002686
RMSE: 11.976941241413376
MAPE: 10.686872819228396
KAMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 23.251670970583447
RMSE: 4.821998648961181
MAPE: 3.833042253232743
MIDPOINT
Prediction vs Close: 51.12% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 16.39872837560197
RMSE: 4.0495343405880595
MAPE: 3.299619771312048
T3
T3([input_arrays], [timeperiod=5], [vfactor=0.7])
Triple Exponential Moving Average (T3) (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 5
vfactor: 0.7
Outputs:
real
19
Working on T3 predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.26 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4414.515, Time=0.02 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3944.062, Time=0.02 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.20 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3715.173, Time=0.04 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3577.471, Time=0.05 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=0.82 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.33 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3579.471, Time=0.11 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 1.849 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1784.736
Date: Sun, 12 Dec 2021 AIC 3577.471
Time: 13:19:29 BIC 3596.235
Sample: 0 HQIC 3584.677
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1982 0.003 -389.844 0.000 -1.204 -1.192
ar.L2 -0.8974 0.006 -139.861 0.000 -0.910 -0.885
ar.L3 -0.3983 0.006 -68.862 0.000 -0.410 -0.387
sigma2 4.9242 0.023 215.469 0.000 4.879 4.969
===================================================================================
Ljung-Box (L1) (Q): 14.55 Jarque-Bera (JB): 2468024.38
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 3.90
Prob(H) (two-sided): 0.00 Kurtosis: 274.15
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04986, saving model to LSTM4.h5
43/43 - 4s - loss: 1.3764 - val_loss: 0.0499 - lr: 0.0010 - 4s/epoch - 92ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.04986 to 0.04710, saving model to LSTM4.h5
43/43 - 0s - loss: 1.2711 - val_loss: 0.0471 - lr: 0.0010 - 216ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04710
43/43 - 0s - loss: 1.1320 - val_loss: 0.0473 - lr: 0.0010 - 209ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.9343 - val_loss: 0.0507 - lr: 0.0010 - 210ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.8210 - val_loss: 0.0539 - lr: 0.0010 - 208ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7722 - val_loss: 0.0568 - lr: 0.0010 - 216ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00007: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7424 - val_loss: 0.0597 - lr: 0.0010 - 194ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7280 - val_loss: 0.0600 - lr: 1.0000e-04 - 196ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7260 - val_loss: 0.0603 - lr: 1.0000e-04 - 203ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7239 - val_loss: 0.0607 - lr: 1.0000e-04 - 210ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7219 - val_loss: 0.0610 - lr: 1.0000e-04 - 198ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00012: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7198 - val_loss: 0.0614 - lr: 1.0000e-04 - 202ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7186 - val_loss: 0.0614 - lr: 1.0000e-05 - 201ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7184 - val_loss: 0.0615 - lr: 1.0000e-05 - 194ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7182 - val_loss: 0.0615 - lr: 1.0000e-05 - 199ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7180 - val_loss: 0.0615 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00017: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7177 - val_loss: 0.0616 - lr: 1.0000e-05 - 199ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7175 - val_loss: 0.0616 - lr: 1.0000e-05 - 214ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7173 - val_loss: 0.0617 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7171 - val_loss: 0.0617 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7169 - val_loss: 0.0618 - lr: 1.0000e-05 - 194ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7166 - val_loss: 0.0618 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7164 - val_loss: 0.0619 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7162 - val_loss: 0.0619 - lr: 1.0000e-05 - 205ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7159 - val_loss: 0.0620 - lr: 1.0000e-05 - 214ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7157 - val_loss: 0.0620 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7155 - val_loss: 0.0621 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7152 - val_loss: 0.0622 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7150 - val_loss: 0.0622 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7147 - val_loss: 0.0623 - lr: 1.0000e-05 - 208ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7145 - val_loss: 0.0623 - lr: 1.0000e-05 - 195ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7143 - val_loss: 0.0624 - lr: 1.0000e-05 - 199ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7140 - val_loss: 0.0625 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7138 - val_loss: 0.0625 - lr: 1.0000e-05 - 200ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7135 - val_loss: 0.0626 - lr: 1.0000e-05 - 208ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7133 - val_loss: 0.0627 - lr: 1.0000e-05 - 191ms/epoch - 4ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7130 - val_loss: 0.0627 - lr: 1.0000e-05 - 212ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7128 - val_loss: 0.0628 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7125 - val_loss: 0.0629 - lr: 1.0000e-05 - 202ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7123 - val_loss: 0.0629 - lr: 1.0000e-05 - 212ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7120 - val_loss: 0.0630 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7118 - val_loss: 0.0631 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7115 - val_loss: 0.0632 - lr: 1.0000e-05 - 204ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7113 - val_loss: 0.0632 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7110 - val_loss: 0.0633 - lr: 1.0000e-05 - 202ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7108 - val_loss: 0.0634 - lr: 1.0000e-05 - 195ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7105 - val_loss: 0.0635 - lr: 1.0000e-05 - 205ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7103 - val_loss: 0.0636 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7100 - val_loss: 0.0636 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7097 - val_loss: 0.0637 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7095 - val_loss: 0.0638 - lr: 1.0000e-05 - 211ms/epoch - 5ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.04710
43/43 - 0s - loss: 0.7092 - val_loss: 0.0639 - lr: 1.0000e-05 - 203ms/epoch - 5ms/step
Epoch 00052: early stopping
SMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 50.37% Accuracy
MSE: 23.31830336349202
RMSE: 4.828902915103183
MAPE: 3.806885992834059
EMA
Prediction vs Close: 55.97% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 31.4391560756623
RMSE: 5.607063052584865
MAPE: 4.398444723456604
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 46.64% Accuracy
MSE: 48.85272948439791
RMSE: 6.989472761546318
MAPE: 5.616901258925532
DEMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 143.4471215002686
RMSE: 11.976941241413376
MAPE: 10.686872819228396
KAMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 23.251670970583447
RMSE: 4.821998648961181
MAPE: 3.833042253232743
MIDPOINT
Prediction vs Close: 51.12% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 16.39872837560197
RMSE: 4.0495343405880595
MAPE: 3.299619771312048
T3
Prediction vs Close: 53.36% Accuracy
Prediction vs Prediction: 48.13% Accuracy
MSE: 85.43536380908036
RMSE: 9.24312521872772
MAPE: 7.5496901284439915
TEMA
TEMA([input_arrays], [timeperiod=30])
Triple Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
9
Working on TEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=inf, Time=0.31 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=4352.703, Time=0.02 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=3889.412, Time=0.02 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=inf, Time=0.15 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=3689.930, Time=0.04 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=3574.245, Time=0.07 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=inf, Time=1.05 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=inf, Time=0.47 sec
ARIMA(3,3,0)(0,0,0)[0] intercept : AIC=3576.245, Time=0.17 sec
Best model: ARIMA(3,3,0)(0,0,0)[0]
Total fit time: 2.305 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 0) Log Likelihood -1783.123
Date: Sun, 12 Dec 2021 AIC 3574.245
Time: 13:20:45 BIC 3593.008
Sample: 0 HQIC 3581.451
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
ar.L1 -1.1480 0.004 -302.430 0.000 -1.155 -1.141
ar.L2 -0.8300 0.008 -99.682 0.000 -0.846 -0.814
ar.L3 -0.3687 0.007 -50.527 0.000 -0.383 -0.354
sigma2 4.9055 0.028 175.970 0.000 4.851 4.960
===================================================================================
Ljung-Box (L1) (Q): 11.61 Jarque-Bera (JB): 1261976.58
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.16 Skew: 2.52
Prob(H) (two-sided): 0.00 Kurtosis: 196.90
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
ARIMA order: (3, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.05138, saving model to LSTM4.h5
90/90 - 4s - loss: 1.3734 - val_loss: 0.0514 - lr: 0.0010 - 4s/epoch - 44ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.05138
90/90 - 0s - loss: 1.2620 - val_loss: 0.0593 - lr: 0.0010 - 414ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.05138
90/90 - 0s - loss: 1.1366 - val_loss: 0.0715 - lr: 0.0010 - 415ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.05138
90/90 - 0s - loss: 1.0216 - val_loss: 0.0842 - lr: 0.0010 - 408ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.9400 - val_loss: 0.0973 - lr: 0.0010 - 424ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8801 - val_loss: 0.1105 - lr: 0.0010 - 420ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8499 - val_loss: 0.1118 - lr: 1.0000e-04 - 404ms/epoch - 4ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8453 - val_loss: 0.1132 - lr: 1.0000e-04 - 404ms/epoch - 4ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8408 - val_loss: 0.1146 - lr: 1.0000e-04 - 401ms/epoch - 4ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8363 - val_loss: 0.1161 - lr: 1.0000e-04 - 414ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8318 - val_loss: 0.1176 - lr: 1.0000e-04 - 397ms/epoch - 4ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8291 - val_loss: 0.1177 - lr: 1.0000e-05 - 383ms/epoch - 4ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8286 - val_loss: 0.1179 - lr: 1.0000e-05 - 384ms/epoch - 4ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8282 - val_loss: 0.1180 - lr: 1.0000e-05 - 410ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8277 - val_loss: 0.1182 - lr: 1.0000e-05 - 410ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8272 - val_loss: 0.1184 - lr: 1.0000e-05 - 396ms/epoch - 4ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8268 - val_loss: 0.1185 - lr: 1.0000e-05 - 401ms/epoch - 4ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8263 - val_loss: 0.1187 - lr: 1.0000e-05 - 408ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8258 - val_loss: 0.1189 - lr: 1.0000e-05 - 397ms/epoch - 4ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8253 - val_loss: 0.1191 - lr: 1.0000e-05 - 402ms/epoch - 4ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8249 - val_loss: 0.1193 - lr: 1.0000e-05 - 394ms/epoch - 4ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8244 - val_loss: 0.1194 - lr: 1.0000e-05 - 391ms/epoch - 4ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8239 - val_loss: 0.1196 - lr: 1.0000e-05 - 401ms/epoch - 4ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8234 - val_loss: 0.1198 - lr: 1.0000e-05 - 398ms/epoch - 4ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8229 - val_loss: 0.1200 - lr: 1.0000e-05 - 415ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8224 - val_loss: 0.1202 - lr: 1.0000e-05 - 414ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8219 - val_loss: 0.1204 - lr: 1.0000e-05 - 398ms/epoch - 4ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8214 - val_loss: 0.1206 - lr: 1.0000e-05 - 407ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8209 - val_loss: 0.1208 - lr: 1.0000e-05 - 392ms/epoch - 4ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8204 - val_loss: 0.1210 - lr: 1.0000e-05 - 398ms/epoch - 4ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8199 - val_loss: 0.1212 - lr: 1.0000e-05 - 392ms/epoch - 4ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8194 - val_loss: 0.1214 - lr: 1.0000e-05 - 403ms/epoch - 4ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8189 - val_loss: 0.1216 - lr: 1.0000e-05 - 410ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8184 - val_loss: 0.1218 - lr: 1.0000e-05 - 399ms/epoch - 4ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8179 - val_loss: 0.1221 - lr: 1.0000e-05 - 404ms/epoch - 4ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8174 - val_loss: 0.1223 - lr: 1.0000e-05 - 401ms/epoch - 4ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8169 - val_loss: 0.1225 - lr: 1.0000e-05 - 406ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8164 - val_loss: 0.1227 - lr: 1.0000e-05 - 405ms/epoch - 4ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8159 - val_loss: 0.1229 - lr: 1.0000e-05 - 393ms/epoch - 4ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8154 - val_loss: 0.1231 - lr: 1.0000e-05 - 410ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8149 - val_loss: 0.1234 - lr: 1.0000e-05 - 381ms/epoch - 4ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8144 - val_loss: 0.1236 - lr: 1.0000e-05 - 392ms/epoch - 4ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8139 - val_loss: 0.1238 - lr: 1.0000e-05 - 393ms/epoch - 4ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8134 - val_loss: 0.1240 - lr: 1.0000e-05 - 413ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8129 - val_loss: 0.1242 - lr: 1.0000e-05 - 417ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8124 - val_loss: 0.1245 - lr: 1.0000e-05 - 401ms/epoch - 4ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8119 - val_loss: 0.1247 - lr: 1.0000e-05 - 413ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8114 - val_loss: 0.1249 - lr: 1.0000e-05 - 385ms/epoch - 4ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8109 - val_loss: 0.1251 - lr: 1.0000e-05 - 394ms/epoch - 4ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8104 - val_loss: 0.1253 - lr: 1.0000e-05 - 393ms/epoch - 4ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.05138
90/90 - 0s - loss: 0.8099 - val_loss: 0.1256 - lr: 1.0000e-05 - 378ms/epoch - 4ms/step
Epoch 00051: early stopping
SMA Prediction vs Close: 55.22% Accuracy Prediction vs Prediction: 50.37% Accuracy MSE: 23.31830336349202 RMSE: 4.828902915103183 MAPE: 3.806885992834059 EMA Prediction vs Close: 55.97% Accuracy Prediction vs Prediction: 47.76% Accuracy MSE: 31.4391560756623 RMSE: 5.607063052584865 MAPE: 4.398444723456604 WMA Prediction vs Close: 52.99% Accuracy Prediction vs Prediction: 46.64% Accuracy MSE: 48.85272948439791 RMSE: 6.989472761546318 MAPE: 5.616901258925532 DEMA Prediction vs Close: 52.99% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 143.4471215002686 RMSE: 11.976941241413376 MAPE: 10.686872819228396 KAMA Prediction vs Close: 54.48% Accuracy Prediction vs Prediction: 49.63% Accuracy MSE: 23.251670970583447 RMSE: 4.821998648961181 MAPE: 3.833042253232743 MIDPOINT Prediction vs Close: 51.12% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 16.39872837560197 RMSE: 4.0495343405880595 MAPE: 3.299619771312048 T3 Prediction vs Close: 53.36% Accuracy Prediction vs Prediction: 48.13% Accuracy MSE: 85.43536380908036 RMSE: 9.24312521872772 MAPE: 7.5496901284439915 TEMA Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 48.88% Accuracy MSE: 17.749245841001986 RMSE: 4.21298538343085 MAPE: 3.636908590169574 Runtime: mins: 9.928002486749998
Architecture Used¶
In [ ]:
from google.colab import files
import cv2
uploaded = files.upload()
Saving Experiment4.png to Experiment4 (1).png
In [ ]:
img = cv2.imread('Experiment4.png')
plt.figure(figsize=(20,10))
plt.axis("off")
plt.title('LSTM Architecture '+imgfile,fontsize=18)
plt.imshow(img)
Out[ ]:
<matplotlib.image.AxesImage at 0x7fda6de16150>
Model Plots¶
In [77]:
with open('simulation4_data.json') as json_file:
simulation4 = json.load(json_file)
fileimg = 'Experiment4'
In [78]:
for i in range(len(list(simulation4.keys()))):
SIM = list(simulation4.keys())[i]
plot_train(simulation4,SIM)
plot_test(simulation4,SIM)
----- Train RMSE for SMA ----- 0.862136599337807 ----- Train_MSE_LSTM for SMA ----- 0.7432795159177583 ----- Train MAE LSTM for SMA ----- 0.6923502813471426
----- Test RMSE for SMA----- 4.828902915103183 ----- Test_MSE_LSTM for SMA----- 23.31830336349202 ----- Test_MAE_LSTM for SMA----- 3.806885992834059
----- Train RMSE for EMA ----- 5.766823380655699 ----- Train_MSE_LSTM for EMA ----- 33.25625190367723 ----- Train MAE LSTM for EMA ----- 5.764868316083851
----- Test RMSE for EMA----- 5.607063052584865 ----- Test_MSE_LSTM for EMA----- 31.4391560756623 ----- Test_MAE_LSTM for EMA----- 4.398444723456604
----- Train RMSE for WMA ----- 5.235584448867116 ----- Train_MSE_LSTM for WMA ----- 27.41134452121919 ----- Train MAE LSTM for WMA ----- 5.227147406870776
----- Test RMSE for WMA----- 6.989472761546318 ----- Test_MSE_LSTM for WMA----- 48.85272948439791 ----- Test_MAE_LSTM for WMA----- 5.616901258925532
----- Train RMSE for DEMA ----- 8.750974100017887 ----- Train_MSE_LSTM for DEMA ----- 76.57954769918385 ----- Train MAE LSTM for DEMA ----- 8.750575766705051
----- Test RMSE for DEMA----- 11.976941241413376 ----- Test_MSE_LSTM for DEMA----- 143.4471215002686 ----- Test_MAE_LSTM for DEMA----- 10.686872819228396
----- Train RMSE for KAMA ----- 2.103880765025744 ----- Train_MSE_LSTM for KAMA ----- 4.42631427344531 ----- Train MAE LSTM for KAMA ----- 2.0681962896101544
----- Test RMSE for KAMA----- 4.821998648961181 ----- Test_MSE_LSTM for KAMA----- 23.251670970583447 ----- Test_MAE_LSTM for KAMA----- 3.833042253232743
----- Train RMSE for MIDPOINT ----- 3.6774127068967175 ----- Train_MSE_LSTM for MIDPOINT ----- 13.523364216845444 ----- Train MAE LSTM for MIDPOINT ----- 3.613733980915334
----- Test RMSE for MIDPOINT----- 4.0495343405880595 ----- Test_MSE_LSTM for MIDPOINT----- 16.39872837560197 ----- Test_MAE_LSTM for MIDPOINT----- 3.299619771312048
----- Train RMSE for T3 ----- 3.104004527969992 ----- Train_MSE_LSTM for T3 ----- 9.634844109658212 ----- Train MAE LSTM for T3 ----- 2.999024202327917
----- Test RMSE for T3----- 9.24312521872772 ----- Test_MSE_LSTM for T3----- 85.43536380908036 ----- Test_MAE_LSTM for T3----- 7.5496901284439915
----- Train RMSE for TEMA ----- 0.9583344967025622 ----- Train_MSE_LSTM for TEMA ----- 0.9184050075701532 ----- Train MAE LSTM for TEMA ----- 0.7602822898638131
----- Test RMSE for TEMA----- 4.21298538343085 ----- Test_MSE_LSTM for TEMA----- 17.749245841001986 ----- Test_MAE_LSTM for TEMA----- 3.636908590169574
Arima w Exogenous Variable Multistep MutiVariate LSTM Hybrid Model Experiment 5¶
In [ ]:
def get_arima_exog(dataframe,original_data, train_len, test_len):
# prepare train and test data for exogenous vr
X_value = pd.DataFrame(low_vol.iloc[:, :])
y_value = pd.DataFrame(low_vol.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
X_scale_dataset = X_scaler.fit_transform(X_value)
y_scale_dataset = y_scaler.fit_transform(y_value)
# Get data and check shape
# X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X_scale_dataset)
y_train, y_test, = split_train_test(y_scale_dataset)
yc_train,yc_test = split_train_test(low_vol_data)
yc = yc_test.values.tolist()
y_train_list = y_train.flatten().tolist()
y_test_list = y_test.flatten().tolist()
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
# Initialize model
model = auto_arima(y_train_list,exogenous = X_train,trace=True, error_action='ignore', start_p=1,start_q=1,max_p=3,max_q=3,d=3,
suppress_warnings=True,stepwise=True,seasonal=True)
# Determine model parameters
print(model.summary())
model.fit(y_train_list,maxiter=200)
order = model.get_params()['order']
print('ARIMA order:', order, '\n')
# Genereate predictions
prediction = []
for i in range(len(y_test_list)):
model = pmdarima.ARIMA(order=order)
model.fit(y_train_list)
# print('working on', i+1, 'of', len(y_test), '-- ' + str(int(100 * (i + 1) / len(y_test))) + '% complete')
prediction.append(model.predict()[0])
y_train_list.append(y_test_list[i])
predictionte = y_scaler.inverse_transform(np.array(prediction).reshape(-1,1))
y_test_ = y_scaler.inverse_transform(np.array(y_test_list).reshape(-1,1))
# Generate error data
mse = mean_squared_error(yc_test, predictionte)
rmse = mse ** 0.5
mae = mean_absolute_error(y_test_ , predictionte )
return yc,predictionte.flatten().tolist(), mse, rmse, mae
In [ ]:
def get_lstm(data,original_data, train_len, test_len,img_file,ma ,lstm_len=3):
# prepare train and test data
X_value = pd.DataFrame(data.iloc[:, :])
y_value = pd.DataFrame(data.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
# Get data and check shape
X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X)
y_train, y_test, = split_train_test(y)
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
det = 20
input_dim = X_train.shape[1]#3
feature_size = X_train.shape[2]#24
output_dim = y_train.shape[1]#1
# Option 1
# Set up & fit LSTM RNN
model = Sequential()
model.add(LSTM(256, activation='relu', kernel_initializer='he_normal', input_shape=(input_dim, feature_size)))
model.add(Dense(units=64,activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(units=output_dim))
model.compile(optimizer=Adam(learning_rate = 0.001), loss='mse')
## Common code
callbacks = [
EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
ModelCheckpoint('LSTM5.h5', verbose=1, save_best_only=True, save_weights_only=True)]
fname1 = img_file+'.png'
tensorflow.keras.utils.plot_model(
model, to_file=fname1, show_shapes=True, show_dtype=False,
show_layer_names=True, expand_nested=False, dpi=96,
layer_range=None, show_layer_activations=False
)
history = model.fit(X_train, y_train, epochs=500, batch_size=int( optimized_period[ma]), verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# plot loss
fname2 = img_file+'-'+ma
plt.title(img_file+'-'+ma+' Loss')
plt.xlabel("Epochs")
plt.ylabel("Loss")
pyplot.plot(history.history['loss'], label='train')
pyplot.plot(history.history['val_loss'], label='validation')
pyplot.legend()
pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# # option 2
# model = Sequential()
# model.add(Bidirectional(LSTM(units= 128), input_shape=(input_dim, feature_size)))
# model.add(Dense(64))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(lr = 0.001), loss='mean_squared_error', metrics=['accuracy'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Option 3
# define custom activation
#
# class Double_Tanh(Activation):
# def __init__(self, activation, **kwargs):
# super(Double_Tanh, self).__init__(activation, **kwargs)
# self.__name__ = 'double_tanh'
# def double_tanh(x):
# return (K.tanh(x) * 2)
# get_custom_objects().update({'double_tanh':Double_Tanh(double_tanh)})
# # Model Generation
# model = Sequential()
# #check https://machinelearningmastery.com/use-weight-regularization-lstm-networks-time-series-forecasting/
# model.add(LSTM(25, input_shape=(input_dim, feature_size), dropout=0.2, kernel_regularizer=l1_l2(0.00,0.00), bias_regularizer=l1_l2(0.00,0.00)))
# model.add(Dense(1))
# model.add(Activation(double_tanh))
# model.compile(loss='mean_squared_error', optimizer='adam', metrics=['mse', 'mae'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Option 4
# Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(units=lstm_len, return_sequences=True, input_shape=(input_dim, feature_size)))
# model.add(LSTM(units=int(lstm_len/2)))
# model.add(Dense(1, activation='sigmoid'))
# model.compile(loss='mean_squared_error', optimizer='adam')
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM5.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Generate predictions
predictiontr = model.predict(X_train, verbose=0)
predictiontr = y_scaler.inverse_transform(predictiontr).tolist()
outputtr = []
for i in range(len(predictiontr)):
outputtr.extend(predictiontr[i])
predictiontr = outputtr
# Generate error data
## replace with yc , xtest generated by new multistep method
mse_tr = mean_squared_error(y_train, predictiontr)
rmse_tr = mse_tr ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictiontr))
mae_tr = mean_absolute_error(y_train, pd.Series(predictiontr))
# Original_tr = pd.Series(yc_train)
Original_tr = y_scaler.inverse_transform(y_train).flatten().tolist()
predictionte = model.predict(X_test, verbose=0)
predictionte = (y_scaler.inverse_transform(predictionte)-det).tolist()
outputte = []
for i in range(len(predictionte)):
outputte.extend(predictionte[i])
predictionte = outputte
# Generate error data
mse_te = mean_squared_error(y_test, predictionte)
rmse_te = mse_te ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictionte))
mae_te = mean_absolute_error(y_test, pd.Series(predictionte))
# Original_te = pd.Series(yc_test)
Original_te = y_scaler.inverse_transform(y_test).flatten().tolist()
return Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,Original_te,predictionte, mse_te,rmse_te,mae_te
In [ ]:
if __name__ == '__main__':
start_time = timeit.default_timer()
simulation5 = {}
imgfile = 'Experiment5'
for ma in optimized_period:
print(ma)
print(functions[ma])
print ( int( optimized_period[ma]))
# if ma == 'SMA':
low_vol = df.apply(lambda c: functions[ma](c, timeperiod = int( optimized_period[ma])))
low_vol = low_vol.fillna(0)
low_vol_data = df['close']
high_vol = pd.DataFrame()
df2 = df.copy()
for i in df2.columns:
if i in low_vol.columns:
high_vol[i] = df2[i].subtract(low_vol[i], fill_value=0)
high_vol_data = df['close']
## *****************************************************
# Generate ARIMA and LSTM predictions
print('\nWorking on ' + ma + ' predictions')
try:
print('parameters used : ', train_len, test_len)
low_vol_Original, low_vol_prediction, low_vol_mse, low_vol_rmse,low_vol_mae = get_arima_exog(low_vol,low_vol_data, train_len, test_len)
except:
print('ARIMA error, skipping to next MA type')
continue
Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,high_vol_Original, high_vol_prediction, high_vol_mse, high_vol_rmse,high_vol_mae, = get_lstm(high_vol,high_vol_data, train_len, test_len,imgfile,ma)
final_prediction_tr = df['close'].head(train_len).values + pd.Series(predictiontr) # ignoring first 3 steps
mse_ftr = mean_squared_error(df['close'].head(train_len).values,final_prediction_tr.values)
rmse_ftr = mse_ftr ** 0.5
mape_ftr = mean_absolute_percentage_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
mae_ftr = mean_absolute_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
final_prediction = pd.Series(low_vol_prediction[3:]) + pd.Series(high_vol_prediction)
mse = mean_squared_error(df['close'].tail(test_len).values,final_prediction.values)
rmse = mse ** 0.5
mape = mean_absolute_percentage_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
mae = mean_absolute_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
# Generate prediction accuracy
actual = df['close'].tail(test_len).values
result_1 = []
result_2 = []
for i in range(1, len(final_prediction)):
# Compare prediction to previous close price
if final_prediction[i] > actual[i-1] and actual[i] > actual[i-1]:
result_1.append(1)
elif final_prediction[i] < actual[i-1] and actual[i] < actual[i-1]:
result_1.append(1)
else:
result_1.append(0)
# Compare prediction to previous prediction
if final_prediction[i] > final_prediction[i-1] and actual[i] > actual[i-1]:
result_2.append(1)
elif final_prediction[i] < final_prediction[i-1] and actual[i] < actual[i-1]:
result_2.append(1)
else:
result_2.append(0)
accuracy_1 = np.mean(result_1)
accuracy_2 = np.mean(result_2)
simulation5[ma] = {'low_vol': {'original':list(low_vol_Original), 'prediction': list(low_vol_prediction) , 'mse': low_vol_mse,
'rmse': low_vol_rmse, 'mae' : low_vol_mae},
'high_vol': {'original':list(high_vol_Original),'prediction': list(high_vol_prediction), 'mse': high_vol_mse,
'rmse': high_vol_rmse, 'mae' : high_vol_mae},
'final_tr': {'original':df['close'].head(train_len).tolist(),'prediction': final_prediction_tr.values.tolist(), 'mse': mse_ftr,
'rmse': rmse_ftr, 'mae' : mae_ftr},
'final': {'original': df['close'].tail(test_len).tolist(), 'prediction': final_prediction.values.tolist(), 'mse': mse,
'rmse': rmse, 'mae': mae },
'accuracy': {'prediction vs close': accuracy_1, 'prediction vs prediction': accuracy_2}}
# save simulation data here as checkpoint
with open('simulation5_data.json', 'w') as fp:
json.dump(simulation5, fp)
for ma in simulation5.keys():
print('\n' + ma)
print('Prediction vs Close:\t\t' + str(round(100*simulation5[ma]['accuracy']['prediction vs close'], 2))
+ '% Accuracy')
print('Prediction vs Prediction:\t' + str(round(100*simulation5[ma]['accuracy']['prediction vs prediction'], 2))
+ '% Accuracy')
print('MSE:\t', simulation5[ma]['final']['mse'],
'\nRMSE:\t', simulation5[ma]['final']['rmse'],
'\nMAPE:\t', simulation5[ma]['final']['mae'])#,
# '\nMAPE:\t', simulation[ma]['final']['mape'])
# else:
# break
elapsed = timeit.default_timer() - start_time
print('Runtime: mins:',elapsed/60)
SMA
SMA([input_arrays], [timeperiod=30])
Simple Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
17
Working on SMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-15000.708, Time=8.38 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-13492.284, Time=2.29 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15827.971, Time=8.04 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-13635.197, Time=10.22 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-14132.778, Time=3.77 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15140.312, Time=10.05 sec
ARIMA(1,3,0)(0,0,0)[0] intercept : AIC=-13970.469, Time=7.25 sec
Best model: ARIMA(1,3,0)(0,0,0)[0]
Total fit time: 50.014 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(1, 3, 0) Log Likelihood 7936.985
Date: Sun, 12 Dec 2021 AIC -15827.971
Time: 13:27:20 BIC -15720.081
Sample: 0 HQIC -15786.537
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -4.786e-05 0.001 -0.066 0.947 -0.001 0.001
x2 -4.789e-05 0.001 -0.085 0.932 -0.001 0.001
x3 -4.819e-05 0.000 -0.105 0.917 -0.001 0.001
x4 1.0000 0.001 1557.248 0.000 0.999 1.001
x5 -4.579e-05 0.001 -0.071 0.943 -0.001 0.001
x6 -5.16e-05 0.000 -0.432 0.666 -0.000 0.000
x7 -4.778e-05 0.000 -0.278 0.781 -0.000 0.000
x8 -0.0012 0.000 -7.403 0.000 -0.002 -0.001
x9 -3.454e-06 0.002 -0.002 0.998 -0.003 0.003
x10 -0.0005 0.001 -0.403 0.687 -0.003 0.002
x11 0.0029 0.000 10.904 0.000 0.002 0.003
x12 -0.0003 0.000 -1.815 0.069 -0.001 2.06e-05
x13 -4.809e-05 0.000 -0.157 0.875 -0.001 0.001
x14 -0.0001 0.000 -0.482 0.630 -0.001 0.000
x15 -5.214e-05 0.000 -0.273 0.785 -0.000 0.000
x16 -4.468e-05 0.000 -0.125 0.901 -0.001 0.001
x17 -4.224e-05 0.000 -0.202 0.840 -0.000 0.000
x18 -8.086e-05 0.000 -0.270 0.787 -0.001 0.001
x19 -5.537e-05 0.000 -0.244 0.807 -0.000 0.000
x20 8.423e-05 0.000 0.333 0.739 -0.000 0.001
x21 -4.232e-05 0.000 -0.166 0.868 -0.001 0.000
ar.L1 -0.6666 6.03e-06 -1.11e+05 0.000 -0.667 -0.667
sigma2 4.093e-10 8.97e-11 4.563 0.000 2.33e-10 5.85e-10
===================================================================================
Ljung-Box (L1) (Q): 60.24 Jarque-Bera (JB): 1334882.31
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.11 Skew: -3.81
Prob(H) (two-sided): 0.00 Kurtosis: 202.35
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 5.73e+20. Standard errors may be unstable.
ARIMA order: (1, 3, 0)
WARNING:tensorflow:Layer lstm_17 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.07596, saving model to LSTM5.h5
48/48 - 2s - loss: 0.4278 - val_loss: 0.0760 - lr: 0.0010 - 2s/epoch - 37ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.07596 to 0.01940, saving model to LSTM5.h5
48/48 - 0s - loss: 0.1376 - val_loss: 0.0194 - lr: 0.0010 - 413ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.01940
48/48 - 0s - loss: 0.0839 - val_loss: 0.8476 - lr: 0.0010 - 359ms/epoch - 7ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.01940
48/48 - 0s - loss: 0.0545 - val_loss: 0.1203 - lr: 0.0010 - 372ms/epoch - 8ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.01940 to 0.01333, saving model to LSTM5.h5
48/48 - 0s - loss: 0.0497 - val_loss: 0.0133 - lr: 0.0010 - 369ms/epoch - 8ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0493 - val_loss: 0.2247 - lr: 0.0010 - 353ms/epoch - 7ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0531 - val_loss: 0.0729 - lr: 0.0010 - 359ms/epoch - 7ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0445 - val_loss: 0.0255 - lr: 0.0010 - 370ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0439 - val_loss: 0.0683 - lr: 0.0010 - 366ms/epoch - 8ms/step
Epoch 10/500
Epoch 00010: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00010: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0365 - val_loss: 0.0208 - lr: 0.0010 - 362ms/epoch - 8ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0370 - val_loss: 0.0241 - lr: 1.0000e-04 - 350ms/epoch - 7ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0370 - val_loss: 0.0294 - lr: 1.0000e-04 - 397ms/epoch - 8ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0369 - val_loss: 0.0328 - lr: 1.0000e-04 - 375ms/epoch - 8ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0400 - val_loss: 0.0344 - lr: 1.0000e-04 - 354ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00015: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0391 - val_loss: 0.0289 - lr: 1.0000e-04 - 403ms/epoch - 8ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0350 - val_loss: 0.0281 - lr: 1.0000e-05 - 349ms/epoch - 7ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0400 - val_loss: 0.0277 - lr: 1.0000e-05 - 369ms/epoch - 8ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0353 - val_loss: 0.0275 - lr: 1.0000e-05 - 346ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0408 - val_loss: 0.0270 - lr: 1.0000e-05 - 359ms/epoch - 7ms/step
Epoch 20/500
Epoch 00020: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00020: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0365 - val_loss: 0.0274 - lr: 1.0000e-05 - 349ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0363 - val_loss: 0.0280 - lr: 1.0000e-05 - 350ms/epoch - 7ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0400 - val_loss: 0.0282 - lr: 1.0000e-05 - 341ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0358 - val_loss: 0.0277 - lr: 1.0000e-05 - 354ms/epoch - 7ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0352 - val_loss: 0.0285 - lr: 1.0000e-05 - 355ms/epoch - 7ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0385 - val_loss: 0.0286 - lr: 1.0000e-05 - 368ms/epoch - 8ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0359 - val_loss: 0.0280 - lr: 1.0000e-05 - 340ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0351 - val_loss: 0.0280 - lr: 1.0000e-05 - 342ms/epoch - 7ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0402 - val_loss: 0.0276 - lr: 1.0000e-05 - 333ms/epoch - 7ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0377 - val_loss: 0.0278 - lr: 1.0000e-05 - 377ms/epoch - 8ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0389 - val_loss: 0.0288 - lr: 1.0000e-05 - 369ms/epoch - 8ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0366 - val_loss: 0.0283 - lr: 1.0000e-05 - 361ms/epoch - 8ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0354 - val_loss: 0.0281 - lr: 1.0000e-05 - 362ms/epoch - 8ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0357 - val_loss: 0.0282 - lr: 1.0000e-05 - 362ms/epoch - 8ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0365 - val_loss: 0.0276 - lr: 1.0000e-05 - 346ms/epoch - 7ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0393 - val_loss: 0.0273 - lr: 1.0000e-05 - 356ms/epoch - 7ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0416 - val_loss: 0.0266 - lr: 1.0000e-05 - 340ms/epoch - 7ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0360 - val_loss: 0.0263 - lr: 1.0000e-05 - 354ms/epoch - 7ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0385 - val_loss: 0.0268 - lr: 1.0000e-05 - 354ms/epoch - 7ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0354 - val_loss: 0.0267 - lr: 1.0000e-05 - 371ms/epoch - 8ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0376 - val_loss: 0.0260 - lr: 1.0000e-05 - 369ms/epoch - 8ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0370 - val_loss: 0.0254 - lr: 1.0000e-05 - 406ms/epoch - 8ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0361 - val_loss: 0.0259 - lr: 1.0000e-05 - 368ms/epoch - 8ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0388 - val_loss: 0.0262 - lr: 1.0000e-05 - 381ms/epoch - 8ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0373 - val_loss: 0.0277 - lr: 1.0000e-05 - 356ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0341 - val_loss: 0.0280 - lr: 1.0000e-05 - 363ms/epoch - 8ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0388 - val_loss: 0.0282 - lr: 1.0000e-05 - 385ms/epoch - 8ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0371 - val_loss: 0.0296 - lr: 1.0000e-05 - 363ms/epoch - 8ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0418 - val_loss: 0.0290 - lr: 1.0000e-05 - 385ms/epoch - 8ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0371 - val_loss: 0.0296 - lr: 1.0000e-05 - 384ms/epoch - 8ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0369 - val_loss: 0.0288 - lr: 1.0000e-05 - 347ms/epoch - 7ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0367 - val_loss: 0.0291 - lr: 1.0000e-05 - 389ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0307 - val_loss: 0.0293 - lr: 1.0000e-05 - 359ms/epoch - 7ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0320 - val_loss: 0.0294 - lr: 1.0000e-05 - 356ms/epoch - 7ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0365 - val_loss: 0.0281 - lr: 1.0000e-05 - 366ms/epoch - 8ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.01333
48/48 - 0s - loss: 0.0369 - val_loss: 0.0275 - lr: 1.0000e-05 - 365ms/epoch - 8ms/step
Epoch 00055: early stopping
SMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.13% Accuracy
MSE: 38.984836670221576
RMSE: 6.24378384236847
MAPE: 5.10393861237253
EMA
EMA([input_arrays], [timeperiod=30])
Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
51
Working on EMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.775, Time=2.37 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=3.89 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16801.081, Time=8.82 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.10 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-14532.068, Time=7.39 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15610.472, Time=11.82 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16103.302, Time=12.30 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17030.021, Time=4.45 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.614, Time=3.05 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.441, Time=5.90 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=17.75 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17003.984, Time=3.22 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-16998.666, Time=3.64 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 89.716 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.720
Date: Sun, 12 Dec 2021 AIC -17087.441
Time: 13:29:53 BIC -16965.479
Sample: 0 HQIC -17040.602
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.316e-10 9.87e-21 -2.35e+10 0.000 -2.32e-10 -2.32e-10
x2 -2.308e-10 9.85e-21 -2.34e+10 0.000 -2.31e-10 -2.31e-10
x3 -2.324e-10 9.88e-21 -2.35e+10 0.000 -2.32e-10 -2.32e-10
x4 1.0000 9.87e-21 1.01e+20 0.000 1.000 1.000
x5 -2.106e-10 9.41e-21 -2.24e+10 0.000 -2.11e-10 -2.11e-10
x6 -7.996e-10 1.74e-20 -4.59e+10 0.000 -8e-10 -8e-10
x7 -2.295e-10 9.82e-21 -2.34e+10 0.000 -2.29e-10 -2.29e-10
x8 -2.244e-10 9.71e-21 -2.31e+10 0.000 -2.24e-10 -2.24e-10
x9 -1.166e-11 1.98e-21 -5.9e+09 0.000 -1.17e-11 -1.17e-11
x10 -4.453e-11 4.22e-21 -1.06e+10 0.000 -4.45e-11 -4.45e-11
x11 -2.219e-10 9.65e-21 -2.3e+10 0.000 -2.22e-10 -2.22e-10
x12 -2.264e-10 9.76e-21 -2.32e+10 0.000 -2.26e-10 -2.26e-10
x13 -2.315e-10 9.87e-21 -2.35e+10 0.000 -2.31e-10 -2.31e-10
x14 -1.766e-09 2.73e-20 -6.48e+10 0.000 -1.77e-09 -1.77e-09
x15 -2.167e-10 9.37e-21 -2.31e+10 0.000 -2.17e-10 -2.17e-10
x16 -5.232e-10 1.49e-20 -3.52e+10 0.000 -5.23e-10 -5.23e-10
x17 -2.147e-10 9.48e-21 -2.27e+10 0.000 -2.15e-10 -2.15e-10
x18 -3.791e-11 3.96e-21 -9.56e+09 0.000 -3.79e-11 -3.79e-11
x19 -2.597e-10 1.05e-20 -2.48e+10 0.000 -2.6e-10 -2.6e-10
x20 -2.417e-10 1e-20 -2.41e+10 0.000 -2.42e-10 -2.42e-10
x21 -4.823e-10 1.4e-20 -3.44e+10 0.000 -4.82e-10 -4.82e-10
ar.L1 -0.4923 1.46e-22 -3.38e+21 0.000 -0.492 -0.492
ar.L2 -0.1923 8.47e-23 -2.27e+21 0.000 -0.192 -0.192
ar.L3 -0.0462 4.02e-23 -1.15e+21 0.000 -0.046 -0.046
ma.L1 -0.7077 3.31e-22 -2.14e+21 0.000 -0.708 -0.708
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 54.09 Jarque-Bera (JB): 4207353.17
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.48
Prob(H) (two-sided): 0.00 Kurtosis: 357.00
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 3.15e+40. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
WARNING:tensorflow:Layer lstm_18 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.37093, saving model to LSTM5.h5
16/16 - 2s - loss: 0.6061 - val_loss: 0.3709 - lr: 0.0010 - 2s/epoch - 100ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.37093 to 0.15384, saving model to LSTM5.h5
16/16 - 0s - loss: 0.2060 - val_loss: 0.1538 - lr: 0.0010 - 155ms/epoch - 10ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.15384 to 0.05870, saving model to LSTM5.h5
16/16 - 0s - loss: 0.0942 - val_loss: 0.0587 - lr: 0.0010 - 154ms/epoch - 10ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.05870 to 0.01500, saving model to LSTM5.h5
16/16 - 0s - loss: 0.0690 - val_loss: 0.0150 - lr: 0.0010 - 144ms/epoch - 9ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.01500 to 0.00667, saving model to LSTM5.h5
16/16 - 0s - loss: 0.0491 - val_loss: 0.0067 - lr: 0.0010 - 172ms/epoch - 11ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0536 - val_loss: 0.0068 - lr: 0.0010 - 137ms/epoch - 9ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0430 - val_loss: 0.0101 - lr: 0.0010 - 137ms/epoch - 9ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0406 - val_loss: 0.0245 - lr: 0.0010 - 133ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0375 - val_loss: 0.0093 - lr: 0.0010 - 139ms/epoch - 9ms/step
Epoch 10/500
Epoch 00010: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00010: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0355 - val_loss: 0.0169 - lr: 0.0010 - 133ms/epoch - 8ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0374 - val_loss: 0.0167 - lr: 1.0000e-04 - 149ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0344 - val_loss: 0.0143 - lr: 1.0000e-04 - 154ms/epoch - 10ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0325 - val_loss: 0.0138 - lr: 1.0000e-04 - 138ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0326 - val_loss: 0.0144 - lr: 1.0000e-04 - 133ms/epoch - 8ms/step
Epoch 15/500
Epoch 00015: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00015: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0288 - val_loss: 0.0146 - lr: 1.0000e-04 - 126ms/epoch - 8ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0315 - val_loss: 0.0145 - lr: 1.0000e-05 - 137ms/epoch - 9ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0359 - val_loss: 0.0145 - lr: 1.0000e-05 - 144ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0350 - val_loss: 0.0146 - lr: 1.0000e-05 - 149ms/epoch - 9ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0296 - val_loss: 0.0147 - lr: 1.0000e-05 - 145ms/epoch - 9ms/step
Epoch 20/500
Epoch 00020: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00020: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0356 - val_loss: 0.0150 - lr: 1.0000e-05 - 150ms/epoch - 9ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0354 - val_loss: 0.0152 - lr: 1.0000e-05 - 132ms/epoch - 8ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0317 - val_loss: 0.0153 - lr: 1.0000e-05 - 127ms/epoch - 8ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0325 - val_loss: 0.0153 - lr: 1.0000e-05 - 128ms/epoch - 8ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0334 - val_loss: 0.0154 - lr: 1.0000e-05 - 141ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0345 - val_loss: 0.0156 - lr: 1.0000e-05 - 156ms/epoch - 10ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0295 - val_loss: 0.0158 - lr: 1.0000e-05 - 136ms/epoch - 8ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0315 - val_loss: 0.0160 - lr: 1.0000e-05 - 142ms/epoch - 9ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0327 - val_loss: 0.0158 - lr: 1.0000e-05 - 134ms/epoch - 8ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0352 - val_loss: 0.0155 - lr: 1.0000e-05 - 143ms/epoch - 9ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0317 - val_loss: 0.0154 - lr: 1.0000e-05 - 136ms/epoch - 8ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0321 - val_loss: 0.0152 - lr: 1.0000e-05 - 140ms/epoch - 9ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0351 - val_loss: 0.0150 - lr: 1.0000e-05 - 130ms/epoch - 8ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0314 - val_loss: 0.0150 - lr: 1.0000e-05 - 144ms/epoch - 9ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0327 - val_loss: 0.0152 - lr: 1.0000e-05 - 129ms/epoch - 8ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0307 - val_loss: 0.0153 - lr: 1.0000e-05 - 126ms/epoch - 8ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0327 - val_loss: 0.0153 - lr: 1.0000e-05 - 137ms/epoch - 9ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0332 - val_loss: 0.0151 - lr: 1.0000e-05 - 126ms/epoch - 8ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0311 - val_loss: 0.0152 - lr: 1.0000e-05 - 124ms/epoch - 8ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0309 - val_loss: 0.0154 - lr: 1.0000e-05 - 131ms/epoch - 8ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0343 - val_loss: 0.0157 - lr: 1.0000e-05 - 141ms/epoch - 9ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0348 - val_loss: 0.0159 - lr: 1.0000e-05 - 139ms/epoch - 9ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0308 - val_loss: 0.0157 - lr: 1.0000e-05 - 143ms/epoch - 9ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0327 - val_loss: 0.0158 - lr: 1.0000e-05 - 145ms/epoch - 9ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0285 - val_loss: 0.0159 - lr: 1.0000e-05 - 126ms/epoch - 8ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0334 - val_loss: 0.0160 - lr: 1.0000e-05 - 135ms/epoch - 8ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0317 - val_loss: 0.0161 - lr: 1.0000e-05 - 133ms/epoch - 8ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0298 - val_loss: 0.0163 - lr: 1.0000e-05 - 131ms/epoch - 8ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0325 - val_loss: 0.0161 - lr: 1.0000e-05 - 132ms/epoch - 8ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0335 - val_loss: 0.0159 - lr: 1.0000e-05 - 162ms/epoch - 10ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0320 - val_loss: 0.0158 - lr: 1.0000e-05 - 134ms/epoch - 8ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0327 - val_loss: 0.0159 - lr: 1.0000e-05 - 133ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0344 - val_loss: 0.0159 - lr: 1.0000e-05 - 144ms/epoch - 9ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0322 - val_loss: 0.0159 - lr: 1.0000e-05 - 128ms/epoch - 8ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0327 - val_loss: 0.0158 - lr: 1.0000e-05 - 136ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00667
16/16 - 0s - loss: 0.0317 - val_loss: 0.0160 - lr: 1.0000e-05 - 132ms/epoch - 8ms/step
Epoch 00055: early stopping
SMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.13% Accuracy
MSE: 38.984836670221576
RMSE: 6.24378384236847
MAPE: 5.10393861237253
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 124.25764226707467
RMSE: 11.1470912020614
MAPE: 9.17724208981177
WMA
WMA([input_arrays], [timeperiod=30])
Weighted Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
49
Working on WMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-14480.432, Time=9.64 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-15747.905, Time=6.37 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15116.389, Time=7.16 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-13532.115, Time=7.90 sec
ARIMA(0,3,0)(0,0,0)[0] intercept : AIC=-13619.624, Time=5.46 sec
Best model: ARIMA(0,3,0)(0,0,0)[0]
Total fit time: 36.552 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(0, 3, 0) Log Likelihood 7895.952
Date: Sun, 12 Dec 2021 AIC -15747.905
Time: 13:38:05 BIC -15644.706
Sample: 0 HQIC -15708.272
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 3.384e-05 1.9e-05 1.778 0.075 -3.47e-06 7.12e-05
x2 3.379e-05 1.84e-05 1.832 0.067 -2.35e-06 6.99e-05
x3 3.388e-05 4.34e-05 0.781 0.435 -5.12e-05 0.000
x4 1.0000 4.12e-06 2.43e+05 0.000 1.000 1.000
x5 3.227e-05 3.52e-06 9.163 0.000 2.54e-05 3.92e-05
x6 5.559e-05 6.75e-05 0.823 0.410 -7.67e-05 0.000
x7 3.369e-05 2.38e-05 1.415 0.157 -1.3e-05 8.03e-05
x8 0.0023 2.6e-05 86.661 0.000 0.002 0.002
x9 -8.72e-06 7.51e-07 -11.610 0.000 -1.02e-05 -7.25e-06
x10 -0.0023 3.33e-05 -67.770 0.000 -0.002 -0.002
x11 0.0093 2.8e-05 333.459 0.000 0.009 0.009
x12 -0.0118 2.37e-05 -498.171 0.000 -0.012 -0.012
x13 3.382e-05 1.49e-05 2.273 0.023 4.66e-06 6.3e-05
x14 9.271e-05 6.21e-05 1.493 0.135 -2.9e-05 0.000
x15 3.096e-05 1.92e-05 1.614 0.106 -6.63e-06 6.86e-05
x16 5.52e-05 7.17e-05 0.770 0.441 -8.53e-05 0.000
x17 3.38e-05 3.2e-05 1.056 0.291 -2.89e-05 9.65e-05
x18 -6.715e-06 8.34e-05 -0.081 0.936 -0.000 0.000
x19 3.428e-05 2.07e-05 1.654 0.098 -6.34e-06 7.49e-05
x20 -8.089e-06 9.55e-05 -0.085 0.933 -0.000 0.000
x21 4.255e-05 0.000 0.094 0.925 -0.001 0.001
sigma2 2.581e-10 7.87e-11 3.280 0.001 1.04e-10 4.12e-10
===================================================================================
Ljung-Box (L1) (Q): 362.92 Jarque-Bera (JB): 5047564.68
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.01 Skew: -11.23
Prob(H) (two-sided): 0.00 Kurtosis: 390.28
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.75e+20. Standard errors may be unstable.
ARIMA order: (0, 3, 0)
WARNING:tensorflow:Layer lstm_19 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 1.75321, saving model to LSTM5.h5
17/17 - 2s - loss: 0.4551 - val_loss: 1.7532 - lr: 0.0010 - 2s/epoch - 115ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 1.75321 to 0.02672, saving model to LSTM5.h5
17/17 - 0s - loss: 0.1938 - val_loss: 0.0267 - lr: 0.0010 - 156ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0837 - val_loss: 0.3487 - lr: 0.0010 - 159ms/epoch - 9ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0603 - val_loss: 0.1463 - lr: 0.0010 - 144ms/epoch - 8ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0479 - val_loss: 0.0528 - lr: 0.0010 - 143ms/epoch - 8ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0416 - val_loss: 0.0410 - lr: 0.0010 - 141ms/epoch - 8ms/step
Epoch 7/500
Epoch 00007: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00007: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0415 - val_loss: 0.0579 - lr: 0.0010 - 138ms/epoch - 8ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0412 - val_loss: 0.0504 - lr: 1.0000e-04 - 138ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0360 - val_loss: 0.0432 - lr: 1.0000e-04 - 141ms/epoch - 8ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0359 - val_loss: 0.0440 - lr: 1.0000e-04 - 155ms/epoch - 9ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0345 - val_loss: 0.0409 - lr: 1.0000e-04 - 148ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00012: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0343 - val_loss: 0.0383 - lr: 1.0000e-04 - 159ms/epoch - 9ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0316 - val_loss: 0.0382 - lr: 1.0000e-05 - 145ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0318 - val_loss: 0.0382 - lr: 1.0000e-05 - 137ms/epoch - 8ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0312 - val_loss: 0.0379 - lr: 1.0000e-05 - 139ms/epoch - 8ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0334 - val_loss: 0.0377 - lr: 1.0000e-05 - 139ms/epoch - 8ms/step
Epoch 17/500
Epoch 00017: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00017: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0312 - val_loss: 0.0373 - lr: 1.0000e-05 - 151ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0325 - val_loss: 0.0371 - lr: 1.0000e-05 - 152ms/epoch - 9ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0324 - val_loss: 0.0372 - lr: 1.0000e-05 - 137ms/epoch - 8ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0319 - val_loss: 0.0370 - lr: 1.0000e-05 - 134ms/epoch - 8ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0316 - val_loss: 0.0369 - lr: 1.0000e-05 - 150ms/epoch - 9ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0336 - val_loss: 0.0364 - lr: 1.0000e-05 - 131ms/epoch - 8ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0322 - val_loss: 0.0364 - lr: 1.0000e-05 - 137ms/epoch - 8ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0319 - val_loss: 0.0360 - lr: 1.0000e-05 - 146ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0311 - val_loss: 0.0357 - lr: 1.0000e-05 - 142ms/epoch - 8ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0323 - val_loss: 0.0354 - lr: 1.0000e-05 - 143ms/epoch - 8ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0327 - val_loss: 0.0349 - lr: 1.0000e-05 - 143ms/epoch - 8ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0338 - val_loss: 0.0342 - lr: 1.0000e-05 - 147ms/epoch - 9ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0311 - val_loss: 0.0344 - lr: 1.0000e-05 - 148ms/epoch - 9ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0354 - val_loss: 0.0344 - lr: 1.0000e-05 - 147ms/epoch - 9ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0364 - val_loss: 0.0342 - lr: 1.0000e-05 - 163ms/epoch - 10ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0345 - val_loss: 0.0344 - lr: 1.0000e-05 - 167ms/epoch - 10ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0340 - val_loss: 0.0343 - lr: 1.0000e-05 - 135ms/epoch - 8ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0299 - val_loss: 0.0344 - lr: 1.0000e-05 - 144ms/epoch - 8ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0321 - val_loss: 0.0339 - lr: 1.0000e-05 - 141ms/epoch - 8ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0285 - val_loss: 0.0336 - lr: 1.0000e-05 - 141ms/epoch - 8ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0362 - val_loss: 0.0335 - lr: 1.0000e-05 - 174ms/epoch - 10ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0316 - val_loss: 0.0335 - lr: 1.0000e-05 - 156ms/epoch - 9ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0310 - val_loss: 0.0331 - lr: 1.0000e-05 - 150ms/epoch - 9ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0299 - val_loss: 0.0324 - lr: 1.0000e-05 - 154ms/epoch - 9ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0324 - val_loss: 0.0320 - lr: 1.0000e-05 - 153ms/epoch - 9ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0334 - val_loss: 0.0330 - lr: 1.0000e-05 - 146ms/epoch - 9ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0317 - val_loss: 0.0336 - lr: 1.0000e-05 - 137ms/epoch - 8ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0321 - val_loss: 0.0332 - lr: 1.0000e-05 - 144ms/epoch - 8ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0312 - val_loss: 0.0333 - lr: 1.0000e-05 - 142ms/epoch - 8ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0297 - val_loss: 0.0339 - lr: 1.0000e-05 - 140ms/epoch - 8ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0323 - val_loss: 0.0335 - lr: 1.0000e-05 - 146ms/epoch - 9ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0329 - val_loss: 0.0340 - lr: 1.0000e-05 - 143ms/epoch - 8ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0332 - val_loss: 0.0345 - lr: 1.0000e-05 - 134ms/epoch - 8ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0326 - val_loss: 0.0354 - lr: 1.0000e-05 - 148ms/epoch - 9ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0283 - val_loss: 0.0355 - lr: 1.0000e-05 - 159ms/epoch - 9ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.02672
17/17 - 0s - loss: 0.0327 - val_loss: 0.0353 - lr: 1.0000e-05 - 159ms/epoch - 9ms/step
Epoch 00052: early stopping
SMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.13% Accuracy
MSE: 38.984836670221576
RMSE: 6.24378384236847
MAPE: 5.10393861237253
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 124.25764226707467
RMSE: 11.1470912020614
MAPE: 9.17724208981177
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 38.21405797999008
RMSE: 6.1817520154071275
MAPE: 5.0592557753421294
DEMA
DEMA([input_arrays], [timeperiod=30])
Double Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
89
Working on DEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.774, Time=2.43 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=4.13 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15590.302, Time=7.08 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.54 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15269.503, Time=8.17 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-16414.961, Time=8.89 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16878.396, Time=9.86 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17030.019, Time=4.40 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.613, Time=3.20 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.441, Time=6.08 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=15.11 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17003.985, Time=3.06 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-16998.665, Time=4.11 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 82.076 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.721
Date: Sun, 12 Dec 2021 AIC -17087.441
Time: 13:40:00 BIC -16965.479
Sample: 0 HQIC -17040.603
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.799e-10 1.43e-20 -1.96e+10 0.000 -2.8e-10 -2.8e-10
x2 -2.817e-10 1.43e-20 -1.97e+10 0.000 -2.82e-10 -2.82e-10
x3 -2.805e-10 1.43e-20 -1.96e+10 0.000 -2.8e-10 -2.8e-10
x4 1.0000 1.43e-20 7e+19 0.000 1.000 1.000
x5 -2.597e-10 1.37e-20 -1.89e+10 0.000 -2.6e-10 -2.6e-10
x6 -1.388e-09 3.12e-20 -4.45e+10 0.000 -1.39e-09 -1.39e-09
x7 -2.789e-10 1.42e-20 -1.96e+10 0.000 -2.79e-10 -2.79e-10
x8 -2.76e-10 1.42e-20 -1.95e+10 0.000 -2.76e-10 -2.76e-10
x9 -2.216e-12 3.53e-22 -6.28e+09 0.000 -2.22e-12 -2.22e-12
x10 -1.345e-10 9.82e-21 -1.37e+10 0.000 -1.34e-10 -1.34e-10
x11 -2.898e-10 1.45e-20 -2e+10 0.000 -2.9e-10 -2.9e-10
x12 -2.602e-10 1.38e-20 -1.89e+10 0.000 -2.6e-10 -2.6e-10
x13 -2.807e-10 1.43e-20 -1.96e+10 0.000 -2.81e-10 -2.81e-10
x14 -1.87e-09 3.69e-20 -5.07e+10 0.000 -1.87e-09 -1.87e-09
x15 -2.726e-10 1.43e-20 -1.9e+10 0.000 -2.73e-10 -2.73e-10
x16 -7.915e-11 7.68e-21 -1.03e+10 0.000 -7.92e-11 -7.92e-11
x17 -2.606e-10 1.33e-20 -1.96e+10 0.000 -2.61e-10 -2.61e-10
x18 -6.408e-10 2.16e-20 -2.97e+10 0.000 -6.41e-10 -6.41e-10
x19 -2.881e-10 1.46e-20 -1.98e+10 0.000 -2.88e-10 -2.88e-10
x20 -4.337e-10 1.78e-20 -2.44e+10 0.000 -4.34e-10 -4.34e-10
x21 -4.549e-10 1.79e-20 -2.55e+10 0.000 -4.55e-10 -4.55e-10
ar.L1 -0.4923 1.46e-22 -3.38e+21 0.000 -0.492 -0.492
ar.L2 -0.1923 8.47e-23 -2.27e+21 0.000 -0.192 -0.192
ar.L3 -0.0461 4.02e-23 -1.15e+21 0.000 -0.046 -0.046
ma.L1 -0.7078 3.31e-22 -2.14e+21 0.000 -0.708 -0.708
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 55.07 Jarque-Bera (JB): 4171695.82
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.26
Prob(H) (two-sided): 0.00 Kurtosis: 355.51
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 2.62e+41. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
WARNING:tensorflow:Layer lstm_20 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04424, saving model to LSTM5.h5
10/10 - 2s - loss: 0.2591 - val_loss: 0.0442 - lr: 0.0010 - 2s/epoch - 154ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.04424
10/10 - 0s - loss: 0.0915 - val_loss: 0.1028 - lr: 0.0010 - 88ms/epoch - 9ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.04424 to 0.02063, saving model to LSTM5.h5
10/10 - 0s - loss: 0.0706 - val_loss: 0.0206 - lr: 0.0010 - 110ms/epoch - 11ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.02063
10/10 - 0s - loss: 0.0560 - val_loss: 0.0392 - lr: 0.0010 - 104ms/epoch - 10ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.02063 to 0.01708, saving model to LSTM5.h5
10/10 - 0s - loss: 0.0487 - val_loss: 0.0171 - lr: 0.0010 - 109ms/epoch - 11ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.01708
10/10 - 0s - loss: 0.0401 - val_loss: 0.0385 - lr: 0.0010 - 98ms/epoch - 10ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.01708
10/10 - 0s - loss: 0.0367 - val_loss: 0.0188 - lr: 0.0010 - 114ms/epoch - 11ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.01708
10/10 - 0s - loss: 0.0320 - val_loss: 0.0195 - lr: 0.0010 - 104ms/epoch - 10ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.01708 to 0.01701, saving model to LSTM5.h5
10/10 - 0s - loss: 0.0301 - val_loss: 0.0170 - lr: 0.0010 - 117ms/epoch - 12ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.01701 to 0.01641, saving model to LSTM5.h5
10/10 - 0s - loss: 0.0302 - val_loss: 0.0164 - lr: 0.0010 - 119ms/epoch - 12ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0299 - val_loss: 0.0195 - lr: 0.0010 - 92ms/epoch - 9ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0243 - val_loss: 0.0169 - lr: 0.0010 - 88ms/epoch - 9ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0264 - val_loss: 0.0206 - lr: 0.0010 - 87ms/epoch - 9ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0269 - val_loss: 0.0205 - lr: 0.0010 - 89ms/epoch - 9ms/step
Epoch 15/500
Epoch 00015: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00015: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0265 - val_loss: 0.0252 - lr: 0.0010 - 91ms/epoch - 9ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0289 - val_loss: 0.0210 - lr: 1.0000e-04 - 88ms/epoch - 9ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0235 - val_loss: 0.0189 - lr: 1.0000e-04 - 95ms/epoch - 9ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0228 - val_loss: 0.0185 - lr: 1.0000e-04 - 101ms/epoch - 10ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0244 - val_loss: 0.0186 - lr: 1.0000e-04 - 99ms/epoch - 10ms/step
Epoch 20/500
Epoch 00020: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00020: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0216 - val_loss: 0.0188 - lr: 1.0000e-04 - 91ms/epoch - 9ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0237 - val_loss: 0.0189 - lr: 1.0000e-05 - 102ms/epoch - 10ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0217 - val_loss: 0.0190 - lr: 1.0000e-05 - 95ms/epoch - 10ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0212 - val_loss: 0.0190 - lr: 1.0000e-05 - 87ms/epoch - 9ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0228 - val_loss: 0.0190 - lr: 1.0000e-05 - 88ms/epoch - 9ms/step
Epoch 25/500
Epoch 00025: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00025: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0227 - val_loss: 0.0190 - lr: 1.0000e-05 - 95ms/epoch - 9ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0232 - val_loss: 0.0190 - lr: 1.0000e-05 - 88ms/epoch - 9ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0223 - val_loss: 0.0190 - lr: 1.0000e-05 - 98ms/epoch - 10ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0212 - val_loss: 0.0190 - lr: 1.0000e-05 - 91ms/epoch - 9ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0229 - val_loss: 0.0190 - lr: 1.0000e-05 - 102ms/epoch - 10ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0224 - val_loss: 0.0190 - lr: 1.0000e-05 - 105ms/epoch - 10ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0225 - val_loss: 0.0190 - lr: 1.0000e-05 - 98ms/epoch - 10ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0221 - val_loss: 0.0190 - lr: 1.0000e-05 - 94ms/epoch - 9ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0224 - val_loss: 0.0191 - lr: 1.0000e-05 - 89ms/epoch - 9ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0243 - val_loss: 0.0191 - lr: 1.0000e-05 - 86ms/epoch - 9ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0232 - val_loss: 0.0191 - lr: 1.0000e-05 - 91ms/epoch - 9ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0231 - val_loss: 0.0191 - lr: 1.0000e-05 - 97ms/epoch - 10ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0226 - val_loss: 0.0191 - lr: 1.0000e-05 - 102ms/epoch - 10ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0248 - val_loss: 0.0191 - lr: 1.0000e-05 - 106ms/epoch - 11ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0226 - val_loss: 0.0191 - lr: 1.0000e-05 - 90ms/epoch - 9ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0253 - val_loss: 0.0191 - lr: 1.0000e-05 - 102ms/epoch - 10ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0244 - val_loss: 0.0190 - lr: 1.0000e-05 - 103ms/epoch - 10ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0229 - val_loss: 0.0189 - lr: 1.0000e-05 - 105ms/epoch - 10ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0237 - val_loss: 0.0189 - lr: 1.0000e-05 - 99ms/epoch - 10ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0239 - val_loss: 0.0188 - lr: 1.0000e-05 - 95ms/epoch - 9ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0230 - val_loss: 0.0187 - lr: 1.0000e-05 - 92ms/epoch - 9ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0217 - val_loss: 0.0187 - lr: 1.0000e-05 - 86ms/epoch - 9ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0233 - val_loss: 0.0188 - lr: 1.0000e-05 - 93ms/epoch - 9ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0259 - val_loss: 0.0188 - lr: 1.0000e-05 - 113ms/epoch - 11ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0232 - val_loss: 0.0188 - lr: 1.0000e-05 - 101ms/epoch - 10ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0226 - val_loss: 0.0188 - lr: 1.0000e-05 - 111ms/epoch - 11ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0219 - val_loss: 0.0187 - lr: 1.0000e-05 - 93ms/epoch - 9ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0226 - val_loss: 0.0187 - lr: 1.0000e-05 - 98ms/epoch - 10ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0225 - val_loss: 0.0186 - lr: 1.0000e-05 - 95ms/epoch - 9ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0225 - val_loss: 0.0187 - lr: 1.0000e-05 - 88ms/epoch - 9ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0265 - val_loss: 0.0187 - lr: 1.0000e-05 - 89ms/epoch - 9ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0218 - val_loss: 0.0188 - lr: 1.0000e-05 - 87ms/epoch - 9ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0216 - val_loss: 0.0188 - lr: 1.0000e-05 - 95ms/epoch - 9ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0230 - val_loss: 0.0188 - lr: 1.0000e-05 - 105ms/epoch - 11ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0216 - val_loss: 0.0189 - lr: 1.0000e-05 - 108ms/epoch - 11ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.01641
10/10 - 0s - loss: 0.0209 - val_loss: 0.0188 - lr: 1.0000e-05 - 94ms/epoch - 9ms/step
Epoch 00060: early stopping
SMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.13% Accuracy
MSE: 38.984836670221576
RMSE: 6.24378384236847
MAPE: 5.10393861237253
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 124.25764226707467
RMSE: 11.1470912020614
MAPE: 9.17724208981177
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 38.21405797999008
RMSE: 6.1817520154071275
MAPE: 5.0592557753421294
DEMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 42.54% Accuracy
MSE: 272.80520892035037
RMSE: 16.51681594376926
MAPE: 15.690440427295842
KAMA
KAMA([input_arrays], [timeperiod=30])
Kaufman Adaptive Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
18
Working on KAMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.902, Time=2.37 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=3.96 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16796.316, Time=8.22 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.31 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-17004.193, Time=2.48 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15176.063, Time=11.13 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16873.638, Time=11.29 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17008.756, Time=2.48 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.764, Time=3.22 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-15723.849, Time=14.02 sec
ARIMA(2,3,0)(0,0,0)[0] intercept : AIC=-17006.756, Time=2.68 sec
Best model: ARIMA(2,3,0)(0,0,0)[0]
Total fit time: 67.184 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(2, 3, 0) Log Likelihood 8528.378
Date: Sun, 12 Dec 2021 AIC -17008.756
Time: 13:49:14 BIC -16896.176
Sample: 0 HQIC -16965.520
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.24e-15 7.41e-26 -3.02e+10 0.000 -2.24e-15 -2.24e-15
x2 8.461e-16 6.6e-26 1.28e+10 0.000 8.46e-16 8.46e-16
x3 4.901e-16 6.89e-26 7.11e+09 0.000 4.9e-16 4.9e-16
x4 1.0000 6.96e-26 1.44e+25 0.000 1.000 1.000
x5 5.931e-15 6.61e-26 8.97e+10 0.000 5.93e-15 5.93e-15
x6 -1.05e-15 1.5e-25 -7e+09 0.000 -1.05e-15 -1.05e-15
x7 1.439e-15 6.87e-26 2.1e+10 0.000 1.44e-15 1.44e-15
x8 -1.25e-15 6.8e-26 -1.84e+10 0.000 -1.25e-15 -1.25e-15
x9 -9.356e-17 8.97e-27 -1.04e+10 0.000 -9.36e-17 -9.36e-17
x10 -1.145e-16 2.88e-26 -3.98e+09 0.000 -1.15e-16 -1.15e-16
x11 -2.036e-16 6.8e-26 -3e+09 0.000 -2.04e-16 -2.04e-16
x12 5.951e-16 6.76e-26 8.81e+09 0.000 5.95e-16 5.95e-16
x13 -6.117e-17 6.94e-26 -8.81e+08 0.000 -6.12e-17 -6.12e-17
x14 1.167e-15 1.99e-25 5.85e+09 0.000 1.17e-15 1.17e-15
x15 -4.274e-14 6.99e-26 -6.11e+11 0.000 -4.27e-14 -4.27e-14
x16 2.262e-14 8.56e-26 2.64e+11 0.000 2.26e-14 2.26e-14
x17 3.384e-14 6.46e-26 5.24e+11 0.000 3.38e-14 3.38e-14
x18 9.894e-16 5.8e-26 1.71e+10 0.000 9.89e-16 9.89e-16
x19 4.115e-14 7.75e-26 5.31e+11 0.000 4.12e-14 4.12e-14
x20 -2.176e-15 9.49e-26 -2.29e+10 0.000 -2.18e-15 -2.18e-15
x21 -7.755e-17 4.63e-26 -1.67e+09 0.000 -7.75e-17 -7.75e-17
ar.L1 -0.9988 9.76e-22 -1.02e+21 0.000 -0.999 -0.999
ar.L2 -0.4972 4.07e-23 -1.22e+22 0.000 -0.497 -0.497
sigma2 1e-10 6.99e-11 1.432 0.152 -3.69e-11 2.37e-10
===================================================================================
Ljung-Box (L1) (Q): 31.54 Jarque-Bera (JB): 2432532.03
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: -0.15
Prob(H) (two-sided): 0.00 Kurtosis: 272.30
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 7.19e+48. Standard errors may be unstable.
ARIMA order: (2, 3, 0)
WARNING:tensorflow:Layer lstm_21 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.01947, saving model to LSTM5.h5
45/45 - 2s - loss: 0.5053 - val_loss: 0.0195 - lr: 0.0010 - 2s/epoch - 39ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.1397 - val_loss: 1.1759 - lr: 0.0010 - 359ms/epoch - 8ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0768 - val_loss: 0.0723 - lr: 0.0010 - 347ms/epoch - 8ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0590 - val_loss: 0.2221 - lr: 0.0010 - 330ms/epoch - 7ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0556 - val_loss: 0.0620 - lr: 0.0010 - 327ms/epoch - 7ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0542 - val_loss: 0.0496 - lr: 0.0010 - 364ms/epoch - 8ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0763 - val_loss: 0.0557 - lr: 1.0000e-04 - 335ms/epoch - 7ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0469 - val_loss: 0.0550 - lr: 1.0000e-04 - 359ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0404 - val_loss: 0.0536 - lr: 1.0000e-04 - 355ms/epoch - 8ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0466 - val_loss: 0.0530 - lr: 1.0000e-04 - 328ms/epoch - 7ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0428 - val_loss: 0.0477 - lr: 1.0000e-04 - 324ms/epoch - 7ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0476 - val_loss: 0.0474 - lr: 1.0000e-05 - 346ms/epoch - 8ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0398 - val_loss: 0.0472 - lr: 1.0000e-05 - 319ms/epoch - 7ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0426 - val_loss: 0.0473 - lr: 1.0000e-05 - 333ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0424 - val_loss: 0.0467 - lr: 1.0000e-05 - 349ms/epoch - 8ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0411 - val_loss: 0.0461 - lr: 1.0000e-05 - 336ms/epoch - 7ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0408 - val_loss: 0.0465 - lr: 1.0000e-05 - 351ms/epoch - 8ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0400 - val_loss: 0.0456 - lr: 1.0000e-05 - 373ms/epoch - 8ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0446 - val_loss: 0.0448 - lr: 1.0000e-05 - 338ms/epoch - 8ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0456 - val_loss: 0.0443 - lr: 1.0000e-05 - 334ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0409 - val_loss: 0.0434 - lr: 1.0000e-05 - 343ms/epoch - 8ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0415 - val_loss: 0.0430 - lr: 1.0000e-05 - 330ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0415 - val_loss: 0.0430 - lr: 1.0000e-05 - 363ms/epoch - 8ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0387 - val_loss: 0.0432 - lr: 1.0000e-05 - 368ms/epoch - 8ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0437 - val_loss: 0.0431 - lr: 1.0000e-05 - 330ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0393 - val_loss: 0.0427 - lr: 1.0000e-05 - 344ms/epoch - 8ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0434 - val_loss: 0.0421 - lr: 1.0000e-05 - 332ms/epoch - 7ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0385 - val_loss: 0.0418 - lr: 1.0000e-05 - 338ms/epoch - 8ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0362 - val_loss: 0.0421 - lr: 1.0000e-05 - 357ms/epoch - 8ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0372 - val_loss: 0.0413 - lr: 1.0000e-05 - 339ms/epoch - 8ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0385 - val_loss: 0.0401 - lr: 1.0000e-05 - 343ms/epoch - 8ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0411 - val_loss: 0.0400 - lr: 1.0000e-05 - 349ms/epoch - 8ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0386 - val_loss: 0.0397 - lr: 1.0000e-05 - 353ms/epoch - 8ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0465 - val_loss: 0.0392 - lr: 1.0000e-05 - 327ms/epoch - 7ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0367 - val_loss: 0.0391 - lr: 1.0000e-05 - 344ms/epoch - 8ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0406 - val_loss: 0.0396 - lr: 1.0000e-05 - 332ms/epoch - 7ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0380 - val_loss: 0.0391 - lr: 1.0000e-05 - 349ms/epoch - 8ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0386 - val_loss: 0.0385 - lr: 1.0000e-05 - 383ms/epoch - 9ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0417 - val_loss: 0.0399 - lr: 1.0000e-05 - 335ms/epoch - 7ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0410 - val_loss: 0.0407 - lr: 1.0000e-05 - 342ms/epoch - 8ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0394 - val_loss: 0.0403 - lr: 1.0000e-05 - 336ms/epoch - 7ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0391 - val_loss: 0.0385 - lr: 1.0000e-05 - 333ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0409 - val_loss: 0.0370 - lr: 1.0000e-05 - 345ms/epoch - 8ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0416 - val_loss: 0.0363 - lr: 1.0000e-05 - 352ms/epoch - 8ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0378 - val_loss: 0.0346 - lr: 1.0000e-05 - 329ms/epoch - 7ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0374 - val_loss: 0.0336 - lr: 1.0000e-05 - 375ms/epoch - 8ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0405 - val_loss: 0.0331 - lr: 1.0000e-05 - 356ms/epoch - 8ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0399 - val_loss: 0.0332 - lr: 1.0000e-05 - 337ms/epoch - 7ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0387 - val_loss: 0.0318 - lr: 1.0000e-05 - 358ms/epoch - 8ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0379 - val_loss: 0.0309 - lr: 1.0000e-05 - 353ms/epoch - 8ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01947
45/45 - 0s - loss: 0.0414 - val_loss: 0.0309 - lr: 1.0000e-05 - 340ms/epoch - 8ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.13% Accuracy
MSE: 38.984836670221576
RMSE: 6.24378384236847
MAPE: 5.10393861237253
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 124.25764226707467
RMSE: 11.1470912020614
MAPE: 9.17724208981177
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 38.21405797999008
RMSE: 6.1817520154071275
MAPE: 5.0592557753421294
DEMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 42.54% Accuracy
MSE: 272.80520892035037
RMSE: 16.51681594376926
MAPE: 15.690440427295842
KAMA
Prediction vs Close: 55.6% Accuracy
Prediction vs Prediction: 45.9% Accuracy
MSE: 39.855043502438484
RMSE: 6.313085101789654
MAPE: 4.932118299391016
MIDPOINT
MIDPOINT([input_arrays], [timeperiod=14])
MidPoint over period (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 14
Outputs:
real
14
Working on MIDPOINT predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.753, Time=2.42 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.592, Time=4.12 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16288.639, Time=11.15 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.592, Time=5.20 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15275.254, Time=7.02 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15486.751, Time=12.72 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=48.000, Time=0.47 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17008.491, Time=2.48 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.554, Time=3.01 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.445, Time=6.05 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=-15686.421, Time=9.57 sec
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17030.168, Time=15.40 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-15138.715, Time=14.83 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 94.480 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.722
Date: Sun, 12 Dec 2021 AIC -17087.445
Time: 13:51:59 BIC -16965.483
Sample: 0 HQIC -17040.607
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.14e-10 1.09e-20 -1.96e+10 0.000 -2.14e-10 -2.14e-10
x2 -2.126e-10 1.13e-20 -1.88e+10 0.000 -2.13e-10 -2.13e-10
x3 -2.175e-10 1.06e-20 -2.04e+10 0.000 -2.17e-10 -2.17e-10
x4 1.0000 1.1e-20 9.11e+19 0.000 1.000 1.000
x5 -1.941e-10 1.05e-20 -1.86e+10 0.000 -1.94e-10 -1.94e-10
x6 -4.131e-09 7.64e-20 -5.4e+10 0.000 -4.13e-09 -4.13e-09
x7 -1.965e-10 1.05e-20 -1.86e+10 0.000 -1.96e-10 -1.96e-10
x8 -1.961e-10 1.07e-20 -1.84e+10 0.000 -1.96e-10 -1.96e-10
x9 -1.005e-10 9.12e-22 -1.1e+11 0.000 -1e-10 -1e-10
x10 -1.739e-10 3.37e-21 -5.16e+10 0.000 -1.74e-10 -1.74e-10
x11 -1.941e-10 1.07e-20 -1.82e+10 0.000 -1.94e-10 -1.94e-10
x12 -2.005e-10 1.06e-20 -1.89e+10 0.000 -2e-10 -2e-10
x13 -2.056e-10 1.07e-20 -1.91e+10 0.000 -2.06e-10 -2.06e-10
x14 -1.687e-09 3.15e-20 -5.36e+10 0.000 -1.69e-09 -1.69e-09
x15 -2.365e-10 1.17e-20 -2.01e+10 0.000 -2.36e-10 -2.36e-10
x16 -1.523e-10 9.42e-21 -1.62e+10 0.000 -1.52e-10 -1.52e-10
x17 -1.491e-10 9.33e-21 -1.6e+10 0.000 -1.49e-10 -1.49e-10
x18 -6.404e-10 1.93e-20 -3.32e+10 0.000 -6.4e-10 -6.4e-10
x19 -2.596e-10 1.23e-20 -2.11e+10 0.000 -2.6e-10 -2.6e-10
x20 -6.246e-10 1.91e-20 -3.28e+10 0.000 -6.25e-10 -6.25e-10
x21 -1.953e-09 2.16e-20 -9.04e+10 0.000 -1.95e-09 -1.95e-09
ar.L1 -0.4914 1.46e-22 -3.38e+21 0.000 -0.491 -0.491
ar.L2 -0.1934 8.48e-23 -2.28e+21 0.000 -0.193 -0.193
ar.L3 -0.0491 4.2e-23 -1.17e+21 0.000 -0.049 -0.049
ma.L1 -0.7092 3.33e-22 -2.13e+21 0.000 -0.709 -0.709
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 32.51 Jarque-Bera (JB): 49038.38
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.35 Skew: 1.06
Prob(H) (two-sided): 0.00 Kurtosis: 41.18
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.71e+40. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
WARNING:tensorflow:Layer lstm_22 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.58849, saving model to LSTM5.h5
58/58 - 2s - loss: 0.2678 - val_loss: 0.5885 - lr: 0.0010 - 2s/epoch - 32ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.58849
58/58 - 0s - loss: 0.2903 - val_loss: 2.1962 - lr: 0.0010 - 431ms/epoch - 7ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.58849
58/58 - 0s - loss: 0.0638 - val_loss: 0.7741 - lr: 0.0010 - 409ms/epoch - 7ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.58849 to 0.30876, saving model to LSTM5.h5
58/58 - 0s - loss: 0.0665 - val_loss: 0.3088 - lr: 0.0010 - 444ms/epoch - 8ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.30876 to 0.13512, saving model to LSTM5.h5
58/58 - 0s - loss: 0.0457 - val_loss: 0.1351 - lr: 0.0010 - 432ms/epoch - 7ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.13512 to 0.06248, saving model to LSTM5.h5
58/58 - 0s - loss: 0.0390 - val_loss: 0.0625 - lr: 0.0010 - 417ms/epoch - 7ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.06248 to 0.03894, saving model to LSTM5.h5
58/58 - 0s - loss: 0.0322 - val_loss: 0.0389 - lr: 0.0010 - 432ms/epoch - 7ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.03894 to 0.03035, saving model to LSTM5.h5
58/58 - 0s - loss: 0.0329 - val_loss: 0.0303 - lr: 0.0010 - 429ms/epoch - 7ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.03035 to 0.01048, saving model to LSTM5.h5
58/58 - 0s - loss: 0.0310 - val_loss: 0.0105 - lr: 0.0010 - 432ms/epoch - 7ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0333 - val_loss: 0.0124 - lr: 0.0010 - 437ms/epoch - 8ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0325 - val_loss: 0.0488 - lr: 0.0010 - 427ms/epoch - 7ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0389 - val_loss: 0.0348 - lr: 0.0010 - 439ms/epoch - 8ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0292 - val_loss: 0.0212 - lr: 0.0010 - 407ms/epoch - 7ms/step
Epoch 14/500
Epoch 00014: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00014: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0331 - val_loss: 0.0954 - lr: 0.0010 - 430ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0299 - val_loss: 0.0724 - lr: 1.0000e-04 - 400ms/epoch - 7ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0231 - val_loss: 0.0653 - lr: 1.0000e-04 - 394ms/epoch - 7ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0251 - val_loss: 0.0666 - lr: 1.0000e-04 - 431ms/epoch - 7ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0246 - val_loss: 0.0709 - lr: 1.0000e-04 - 429ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00019: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0267 - val_loss: 0.0765 - lr: 1.0000e-04 - 426ms/epoch - 7ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0249 - val_loss: 0.0769 - lr: 1.0000e-05 - 412ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0241 - val_loss: 0.0792 - lr: 1.0000e-05 - 450ms/epoch - 8ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0232 - val_loss: 0.0794 - lr: 1.0000e-05 - 420ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0243 - val_loss: 0.0797 - lr: 1.0000e-05 - 400ms/epoch - 7ms/step
Epoch 24/500
Epoch 00024: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00024: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0244 - val_loss: 0.0789 - lr: 1.0000e-05 - 418ms/epoch - 7ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0249 - val_loss: 0.0781 - lr: 1.0000e-05 - 425ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0234 - val_loss: 0.0779 - lr: 1.0000e-05 - 419ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0221 - val_loss: 0.0786 - lr: 1.0000e-05 - 394ms/epoch - 7ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0252 - val_loss: 0.0784 - lr: 1.0000e-05 - 442ms/epoch - 8ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0226 - val_loss: 0.0797 - lr: 1.0000e-05 - 386ms/epoch - 7ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0226 - val_loss: 0.0810 - lr: 1.0000e-05 - 440ms/epoch - 8ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0231 - val_loss: 0.0806 - lr: 1.0000e-05 - 421ms/epoch - 7ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0259 - val_loss: 0.0785 - lr: 1.0000e-05 - 433ms/epoch - 7ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0246 - val_loss: 0.0774 - lr: 1.0000e-05 - 425ms/epoch - 7ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0238 - val_loss: 0.0785 - lr: 1.0000e-05 - 413ms/epoch - 7ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0205 - val_loss: 0.0778 - lr: 1.0000e-05 - 443ms/epoch - 8ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0248 - val_loss: 0.0806 - lr: 1.0000e-05 - 409ms/epoch - 7ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0212 - val_loss: 0.0810 - lr: 1.0000e-05 - 403ms/epoch - 7ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0225 - val_loss: 0.0823 - lr: 1.0000e-05 - 402ms/epoch - 7ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0240 - val_loss: 0.0832 - lr: 1.0000e-05 - 418ms/epoch - 7ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0246 - val_loss: 0.0826 - lr: 1.0000e-05 - 412ms/epoch - 7ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0214 - val_loss: 0.0823 - lr: 1.0000e-05 - 417ms/epoch - 7ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0248 - val_loss: 0.0821 - lr: 1.0000e-05 - 401ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0231 - val_loss: 0.0841 - lr: 1.0000e-05 - 398ms/epoch - 7ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0239 - val_loss: 0.0841 - lr: 1.0000e-05 - 406ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0232 - val_loss: 0.0856 - lr: 1.0000e-05 - 449ms/epoch - 8ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0259 - val_loss: 0.0860 - lr: 1.0000e-05 - 434ms/epoch - 7ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0229 - val_loss: 0.0873 - lr: 1.0000e-05 - 425ms/epoch - 7ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0234 - val_loss: 0.0898 - lr: 1.0000e-05 - 402ms/epoch - 7ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0221 - val_loss: 0.0903 - lr: 1.0000e-05 - 442ms/epoch - 8ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0222 - val_loss: 0.0909 - lr: 1.0000e-05 - 444ms/epoch - 8ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0220 - val_loss: 0.0907 - lr: 1.0000e-05 - 418ms/epoch - 7ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0205 - val_loss: 0.0895 - lr: 1.0000e-05 - 423ms/epoch - 7ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0212 - val_loss: 0.0882 - lr: 1.0000e-05 - 409ms/epoch - 7ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0238 - val_loss: 0.0899 - lr: 1.0000e-05 - 411ms/epoch - 7ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0205 - val_loss: 0.0885 - lr: 1.0000e-05 - 415ms/epoch - 7ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0222 - val_loss: 0.0884 - lr: 1.0000e-05 - 407ms/epoch - 7ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0239 - val_loss: 0.0912 - lr: 1.0000e-05 - 410ms/epoch - 7ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0228 - val_loss: 0.0936 - lr: 1.0000e-05 - 429ms/epoch - 7ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.01048
58/58 - 0s - loss: 0.0235 - val_loss: 0.0936 - lr: 1.0000e-05 - 449ms/epoch - 8ms/step
Epoch 00059: early stopping
SMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.13% Accuracy
MSE: 38.984836670221576
RMSE: 6.24378384236847
MAPE: 5.10393861237253
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 124.25764226707467
RMSE: 11.1470912020614
MAPE: 9.17724208981177
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 38.21405797999008
RMSE: 6.1817520154071275
MAPE: 5.0592557753421294
DEMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 42.54% Accuracy
MSE: 272.80520892035037
RMSE: 16.51681594376926
MAPE: 15.690440427295842
KAMA
Prediction vs Close: 55.6% Accuracy
Prediction vs Prediction: 45.9% Accuracy
MSE: 39.855043502438484
RMSE: 6.313085101789654
MAPE: 4.932118299391016
MIDPOINT
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 47.39% Accuracy
MSE: 247.8059617737088
RMSE: 15.741853822650901
MAPE: 13.137429929578502
T3
T3([input_arrays], [timeperiod=5], [vfactor=0.7])
Triple Exponential Moving Average (T3) (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 5
vfactor: 0.7
Outputs:
real
19
Working on T3 predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-16569.270, Time=2.41 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14511.291, Time=2.44 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15408.738, Time=7.86 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-15165.005, Time=7.88 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15595.465, Time=7.24 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15837.470, Time=9.75 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-15491.538, Time=9.88 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-16378.438, Time=2.47 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-16318.604, Time=3.35 sec
ARIMA(1,3,1)(0,0,0)[0] intercept : AIC=-16567.270, Time=2.21 sec
Best model: ARIMA(1,3,1)(0,0,0)[0]
Total fit time: 55.514 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(1, 3, 1) Log Likelihood 8308.635
Date: Sun, 12 Dec 2021 AIC -16569.270
Time: 14:00:45 BIC -16456.690
Sample: 0 HQIC -16526.035
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 1.355e-13 3.43e-05 3.95e-09 1.000 -6.72e-05 6.72e-05
x2 5.009e-14 2.67e-05 1.88e-09 1.000 -5.23e-05 5.23e-05
x3 -9.101e-15 2.19e-05 -4.16e-10 1.000 -4.29e-05 4.29e-05
x4 1.0000 2.97e-05 3.37e+04 0.000 1.000 1.000
x5 3.626e-12 3.2e-05 1.13e-07 1.000 -6.28e-05 6.28e-05
x6 6.879e-17 0.000 5.13e-13 1.000 -0.000 0.000
x7 1.588e-13 4.04e-05 3.93e-09 1.000 -7.92e-05 7.92e-05
x8 -0.0002 9.77e-06 -20.395 0.000 -0.000 -0.000
x9 3.877e-14 0.001 6.24e-11 1.000 -0.001 0.001
x10 -7.41e-05 0.001 -0.129 0.897 -0.001 0.001
x11 0.0003 4.91e-05 6.926 0.000 0.000 0.000
x12 -0.0004 7.27e-05 -5.556 0.000 -0.001 -0.000
x13 -2.679e-14 3.39e-05 -7.9e-10 1.000 -6.65e-05 6.65e-05
x14 2.97e-13 0.000 2.31e-09 1.000 -0.000 0.000
x15 1.602e-12 7.47e-05 2.14e-08 1.000 -0.000 0.000
x16 -8.756e-13 4.29e-05 -2.04e-08 1.000 -8.41e-05 8.41e-05
x17 1.793e-12 6.56e-05 2.74e-08 1.000 -0.000 0.000
x18 -1.019e-13 0.000 -5.54e-10 1.000 -0.000 0.000
x19 -1.077e-12 8.29e-05 -1.3e-08 1.000 -0.000 0.000
x20 1.771e-13 8.45e-05 2.1e-09 1.000 -0.000 0.000
x21 9.233e-16 0.000 1.94e-12 1.000 -0.001 0.001
ar.L1 -0.2857 0.000 -2747.572 0.000 -0.286 -0.285
ma.L1 -0.9142 7.12e-06 -1.28e+05 0.000 -0.914 -0.914
sigma2 1e-10 7e-11 1.429 0.153 -3.71e-11 2.37e-10
===================================================================================
Ljung-Box (L1) (Q): 84.32 Jarque-Bera (JB): 4804295.53
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.87
Prob(H) (two-sided): 0.00 Kurtosis: 381.28
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 2.06e+20. Standard errors may be unstable.
ARIMA order: (1, 3, 1)
WARNING:tensorflow:Layer lstm_23 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.16122, saving model to LSTM5.h5
43/43 - 2s - loss: 0.2956 - val_loss: 0.1612 - lr: 0.0010 - 2s/epoch - 42ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.16122 to 0.02401, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0668 - val_loss: 0.0240 - lr: 0.0010 - 333ms/epoch - 8ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.02401
43/43 - 0s - loss: 0.0778 - val_loss: 0.1607 - lr: 0.0010 - 315ms/epoch - 7ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.02401
43/43 - 0s - loss: 0.0580 - val_loss: 0.4810 - lr: 0.0010 - 296ms/epoch - 7ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.02401
43/43 - 0s - loss: 0.0439 - val_loss: 0.0782 - lr: 0.0010 - 319ms/epoch - 7ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.02401 to 0.01150, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0390 - val_loss: 0.0115 - lr: 0.0010 - 333ms/epoch - 8ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.01150 to 0.00908, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0450 - val_loss: 0.0091 - lr: 0.0010 - 345ms/epoch - 8ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0358 - val_loss: 0.0364 - lr: 0.0010 - 327ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0299 - val_loss: 0.0156 - lr: 0.0010 - 333ms/epoch - 8ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0331 - val_loss: 0.0099 - lr: 0.0010 - 312ms/epoch - 7ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0290 - val_loss: 0.0154 - lr: 0.0010 - 344ms/epoch - 8ms/step
Epoch 12/500
Epoch 00012: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00012: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0308 - val_loss: 0.1072 - lr: 0.0010 - 321ms/epoch - 7ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0293 - val_loss: 0.0905 - lr: 1.0000e-04 - 305ms/epoch - 7ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0274 - val_loss: 0.0716 - lr: 1.0000e-04 - 317ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0280 - val_loss: 0.0534 - lr: 1.0000e-04 - 303ms/epoch - 7ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0279 - val_loss: 0.0413 - lr: 1.0000e-04 - 308ms/epoch - 7ms/step
Epoch 17/500
Epoch 00017: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00017: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0267 - val_loss: 0.0300 - lr: 1.0000e-04 - 339ms/epoch - 8ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0270 - val_loss: 0.0291 - lr: 1.0000e-05 - 300ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0269 - val_loss: 0.0282 - lr: 1.0000e-05 - 300ms/epoch - 7ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0254 - val_loss: 0.0272 - lr: 1.0000e-05 - 309ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0274 - val_loss: 0.0262 - lr: 1.0000e-05 - 300ms/epoch - 7ms/step
Epoch 22/500
Epoch 00022: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00022: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0267 - val_loss: 0.0252 - lr: 1.0000e-05 - 298ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0241 - val_loss: 0.0243 - lr: 1.0000e-05 - 317ms/epoch - 7ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0258 - val_loss: 0.0235 - lr: 1.0000e-05 - 346ms/epoch - 8ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0274 - val_loss: 0.0226 - lr: 1.0000e-05 - 305ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0269 - val_loss: 0.0217 - lr: 1.0000e-05 - 302ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0255 - val_loss: 0.0207 - lr: 1.0000e-05 - 310ms/epoch - 7ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0270 - val_loss: 0.0200 - lr: 1.0000e-05 - 301ms/epoch - 7ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0254 - val_loss: 0.0192 - lr: 1.0000e-05 - 310ms/epoch - 7ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0264 - val_loss: 0.0185 - lr: 1.0000e-05 - 305ms/epoch - 7ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0267 - val_loss: 0.0178 - lr: 1.0000e-05 - 312ms/epoch - 7ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0251 - val_loss: 0.0174 - lr: 1.0000e-05 - 323ms/epoch - 8ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0266 - val_loss: 0.0167 - lr: 1.0000e-05 - 317ms/epoch - 7ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0249 - val_loss: 0.0160 - lr: 1.0000e-05 - 332ms/epoch - 8ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0261 - val_loss: 0.0155 - lr: 1.0000e-05 - 315ms/epoch - 7ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0269 - val_loss: 0.0149 - lr: 1.0000e-05 - 319ms/epoch - 7ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0244 - val_loss: 0.0144 - lr: 1.0000e-05 - 321ms/epoch - 7ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0261 - val_loss: 0.0139 - lr: 1.0000e-05 - 331ms/epoch - 8ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0233 - val_loss: 0.0136 - lr: 1.0000e-05 - 306ms/epoch - 7ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0248 - val_loss: 0.0132 - lr: 1.0000e-05 - 333ms/epoch - 8ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0272 - val_loss: 0.0127 - lr: 1.0000e-05 - 347ms/epoch - 8ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0263 - val_loss: 0.0122 - lr: 1.0000e-05 - 306ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0259 - val_loss: 0.0117 - lr: 1.0000e-05 - 325ms/epoch - 8ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0245 - val_loss: 0.0111 - lr: 1.0000e-05 - 312ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0265 - val_loss: 0.0107 - lr: 1.0000e-05 - 299ms/epoch - 7ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0248 - val_loss: 0.0104 - lr: 1.0000e-05 - 333ms/epoch - 8ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0276 - val_loss: 0.0102 - lr: 1.0000e-05 - 316ms/epoch - 7ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0274 - val_loss: 0.0098 - lr: 1.0000e-05 - 356ms/epoch - 8ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0242 - val_loss: 0.0094 - lr: 1.0000e-05 - 313ms/epoch - 7ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0258 - val_loss: 0.0092 - lr: 1.0000e-05 - 329ms/epoch - 8ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00908
43/43 - 0s - loss: 0.0251 - val_loss: 0.0092 - lr: 1.0000e-05 - 326ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss improved from 0.00908 to 0.00904, saving model to LSTM5.h5
43/43 - 1s - loss: 0.0269 - val_loss: 0.0090 - lr: 1.0000e-05 - 528ms/epoch - 12ms/step
Epoch 53/500
Epoch 00053: val_loss improved from 0.00904 to 0.00881, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0265 - val_loss: 0.0088 - lr: 1.0000e-05 - 334ms/epoch - 8ms/step
Epoch 54/500
Epoch 00054: val_loss improved from 0.00881 to 0.00874, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0261 - val_loss: 0.0087 - lr: 1.0000e-05 - 356ms/epoch - 8ms/step
Epoch 55/500
Epoch 00055: val_loss improved from 0.00874 to 0.00866, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0257 - val_loss: 0.0087 - lr: 1.0000e-05 - 343ms/epoch - 8ms/step
Epoch 56/500
Epoch 00056: val_loss improved from 0.00866 to 0.00844, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0222 - val_loss: 0.0084 - lr: 1.0000e-05 - 365ms/epoch - 8ms/step
Epoch 57/500
Epoch 00057: val_loss improved from 0.00844 to 0.00830, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0230 - val_loss: 0.0083 - lr: 1.0000e-05 - 322ms/epoch - 7ms/step
Epoch 58/500
Epoch 00058: val_loss improved from 0.00830 to 0.00809, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0262 - val_loss: 0.0081 - lr: 1.0000e-05 - 353ms/epoch - 8ms/step
Epoch 59/500
Epoch 00059: val_loss improved from 0.00809 to 0.00801, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0251 - val_loss: 0.0080 - lr: 1.0000e-05 - 344ms/epoch - 8ms/step
Epoch 60/500
Epoch 00060: val_loss improved from 0.00801 to 0.00799, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0233 - val_loss: 0.0080 - lr: 1.0000e-05 - 339ms/epoch - 8ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.00799
43/43 - 0s - loss: 0.0235 - val_loss: 0.0080 - lr: 1.0000e-05 - 349ms/epoch - 8ms/step
Epoch 62/500
Epoch 00062: val_loss improved from 0.00799 to 0.00799, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0249 - val_loss: 0.0080 - lr: 1.0000e-05 - 357ms/epoch - 8ms/step
Epoch 63/500
Epoch 00063: val_loss improved from 0.00799 to 0.00797, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0250 - val_loss: 0.0080 - lr: 1.0000e-05 - 349ms/epoch - 8ms/step
Epoch 64/500
Epoch 00064: val_loss improved from 0.00797 to 0.00788, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0255 - val_loss: 0.0079 - lr: 1.0000e-05 - 354ms/epoch - 8ms/step
Epoch 65/500
Epoch 00065: val_loss improved from 0.00788 to 0.00785, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0252 - val_loss: 0.0079 - lr: 1.0000e-05 - 346ms/epoch - 8ms/step
Epoch 66/500
Epoch 00066: val_loss improved from 0.00785 to 0.00785, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0241 - val_loss: 0.0078 - lr: 1.0000e-05 - 368ms/epoch - 9ms/step
Epoch 67/500
Epoch 00067: val_loss improved from 0.00785 to 0.00784, saving model to LSTM5.h5
43/43 - 0s - loss: 0.0223 - val_loss: 0.0078 - lr: 1.0000e-05 - 374ms/epoch - 9ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0253 - val_loss: 0.0079 - lr: 1.0000e-05 - 320ms/epoch - 7ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0263 - val_loss: 0.0079 - lr: 1.0000e-05 - 332ms/epoch - 8ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0258 - val_loss: 0.0080 - lr: 1.0000e-05 - 344ms/epoch - 8ms/step
Epoch 71/500
Epoch 00071: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0249 - val_loss: 0.0081 - lr: 1.0000e-05 - 313ms/epoch - 7ms/step
Epoch 72/500
Epoch 00072: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0263 - val_loss: 0.0081 - lr: 1.0000e-05 - 314ms/epoch - 7ms/step
Epoch 73/500
Epoch 00073: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0250 - val_loss: 0.0081 - lr: 1.0000e-05 - 339ms/epoch - 8ms/step
Epoch 74/500
Epoch 00074: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0259 - val_loss: 0.0082 - lr: 1.0000e-05 - 325ms/epoch - 8ms/step
Epoch 75/500
Epoch 00075: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0256 - val_loss: 0.0083 - lr: 1.0000e-05 - 321ms/epoch - 7ms/step
Epoch 76/500
Epoch 00076: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0229 - val_loss: 0.0083 - lr: 1.0000e-05 - 331ms/epoch - 8ms/step
Epoch 77/500
Epoch 00077: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0228 - val_loss: 0.0083 - lr: 1.0000e-05 - 306ms/epoch - 7ms/step
Epoch 78/500
Epoch 00078: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0237 - val_loss: 0.0085 - lr: 1.0000e-05 - 307ms/epoch - 7ms/step
Epoch 79/500
Epoch 00079: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0259 - val_loss: 0.0086 - lr: 1.0000e-05 - 319ms/epoch - 7ms/step
Epoch 80/500
Epoch 00080: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0255 - val_loss: 0.0087 - lr: 1.0000e-05 - 313ms/epoch - 7ms/step
Epoch 81/500
Epoch 00081: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0241 - val_loss: 0.0085 - lr: 1.0000e-05 - 314ms/epoch - 7ms/step
Epoch 82/500
Epoch 00082: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0238 - val_loss: 0.0084 - lr: 1.0000e-05 - 336ms/epoch - 8ms/step
Epoch 83/500
Epoch 00083: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0242 - val_loss: 0.0082 - lr: 1.0000e-05 - 371ms/epoch - 9ms/step
Epoch 84/500
Epoch 00084: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0265 - val_loss: 0.0082 - lr: 1.0000e-05 - 354ms/epoch - 8ms/step
Epoch 85/500
Epoch 00085: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0250 - val_loss: 0.0083 - lr: 1.0000e-05 - 335ms/epoch - 8ms/step
Epoch 86/500
Epoch 00086: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0272 - val_loss: 0.0084 - lr: 1.0000e-05 - 337ms/epoch - 8ms/step
Epoch 87/500
Epoch 00087: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0254 - val_loss: 0.0085 - lr: 1.0000e-05 - 342ms/epoch - 8ms/step
Epoch 88/500
Epoch 00088: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0257 - val_loss: 0.0086 - lr: 1.0000e-05 - 324ms/epoch - 8ms/step
Epoch 89/500
Epoch 00089: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0254 - val_loss: 0.0086 - lr: 1.0000e-05 - 326ms/epoch - 8ms/step
Epoch 90/500
Epoch 00090: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0218 - val_loss: 0.0087 - lr: 1.0000e-05 - 310ms/epoch - 7ms/step
Epoch 91/500
Epoch 00091: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0226 - val_loss: 0.0089 - lr: 1.0000e-05 - 335ms/epoch - 8ms/step
Epoch 92/500
Epoch 00092: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0240 - val_loss: 0.0088 - lr: 1.0000e-05 - 311ms/epoch - 7ms/step
Epoch 93/500
Epoch 00093: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0248 - val_loss: 0.0087 - lr: 1.0000e-05 - 348ms/epoch - 8ms/step
Epoch 94/500
Epoch 00094: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0228 - val_loss: 0.0084 - lr: 1.0000e-05 - 360ms/epoch - 8ms/step
Epoch 95/500
Epoch 00095: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0239 - val_loss: 0.0083 - lr: 1.0000e-05 - 333ms/epoch - 8ms/step
Epoch 96/500
Epoch 00096: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0223 - val_loss: 0.0083 - lr: 1.0000e-05 - 338ms/epoch - 8ms/step
Epoch 97/500
Epoch 00097: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0270 - val_loss: 0.0082 - lr: 1.0000e-05 - 350ms/epoch - 8ms/step
Epoch 98/500
Epoch 00098: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0229 - val_loss: 0.0084 - lr: 1.0000e-05 - 302ms/epoch - 7ms/step
Epoch 99/500
Epoch 00099: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0272 - val_loss: 0.0084 - lr: 1.0000e-05 - 363ms/epoch - 8ms/step
Epoch 100/500
Epoch 00100: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0250 - val_loss: 0.0086 - lr: 1.0000e-05 - 338ms/epoch - 8ms/step
Epoch 101/500
Epoch 00101: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0249 - val_loss: 0.0087 - lr: 1.0000e-05 - 329ms/epoch - 8ms/step
Epoch 102/500
Epoch 00102: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0250 - val_loss: 0.0083 - lr: 1.0000e-05 - 305ms/epoch - 7ms/step
Epoch 103/500
Epoch 00103: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0273 - val_loss: 0.0082 - lr: 1.0000e-05 - 341ms/epoch - 8ms/step
Epoch 104/500
Epoch 00104: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0241 - val_loss: 0.0083 - lr: 1.0000e-05 - 298ms/epoch - 7ms/step
Epoch 105/500
Epoch 00105: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0234 - val_loss: 0.0083 - lr: 1.0000e-05 - 334ms/epoch - 8ms/step
Epoch 106/500
Epoch 00106: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0241 - val_loss: 0.0084 - lr: 1.0000e-05 - 347ms/epoch - 8ms/step
Epoch 107/500
Epoch 00107: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0256 - val_loss: 0.0083 - lr: 1.0000e-05 - 317ms/epoch - 7ms/step
Epoch 108/500
Epoch 00108: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0228 - val_loss: 0.0085 - lr: 1.0000e-05 - 325ms/epoch - 8ms/step
Epoch 109/500
Epoch 00109: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0241 - val_loss: 0.0083 - lr: 1.0000e-05 - 350ms/epoch - 8ms/step
Epoch 110/500
Epoch 00110: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0251 - val_loss: 0.0085 - lr: 1.0000e-05 - 308ms/epoch - 7ms/step
Epoch 111/500
Epoch 00111: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0242 - val_loss: 0.0085 - lr: 1.0000e-05 - 312ms/epoch - 7ms/step
Epoch 112/500
Epoch 00112: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0226 - val_loss: 0.0084 - lr: 1.0000e-05 - 334ms/epoch - 8ms/step
Epoch 113/500
Epoch 00113: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0255 - val_loss: 0.0085 - lr: 1.0000e-05 - 323ms/epoch - 8ms/step
Epoch 114/500
Epoch 00114: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0232 - val_loss: 0.0084 - lr: 1.0000e-05 - 314ms/epoch - 7ms/step
Epoch 115/500
Epoch 00115: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0253 - val_loss: 0.0085 - lr: 1.0000e-05 - 333ms/epoch - 8ms/step
Epoch 116/500
Epoch 00116: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0244 - val_loss: 0.0086 - lr: 1.0000e-05 - 353ms/epoch - 8ms/step
Epoch 117/500
Epoch 00117: val_loss did not improve from 0.00784
43/43 - 0s - loss: 0.0213 - val_loss: 0.0088 - lr: 1.0000e-05 - 320ms/epoch - 7ms/step
Epoch 00117: early stopping
SMA
Prediction vs Close: 51.49% Accuracy
Prediction vs Prediction: 48.13% Accuracy
MSE: 38.984836670221576
RMSE: 6.24378384236847
MAPE: 5.10393861237253
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 124.25764226707467
RMSE: 11.1470912020614
MAPE: 9.17724208981177
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 38.21405797999008
RMSE: 6.1817520154071275
MAPE: 5.0592557753421294
DEMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 42.54% Accuracy
MSE: 272.80520892035037
RMSE: 16.51681594376926
MAPE: 15.690440427295842
KAMA
Prediction vs Close: 55.6% Accuracy
Prediction vs Prediction: 45.9% Accuracy
MSE: 39.855043502438484
RMSE: 6.313085101789654
MAPE: 4.932118299391016
MIDPOINT
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 47.39% Accuracy
MSE: 247.8059617737088
RMSE: 15.741853822650901
MAPE: 13.137429929578502
T3
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 42.91% Accuracy
MSE: 210.84512917819418
RMSE: 14.520507194247527
MAPE: 11.877711162377306
TEMA
TEMA([input_arrays], [timeperiod=30])
Triple Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
9
Working on TEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-16493.570, Time=2.44 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-15527.581, Time=7.40 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16154.477, Time=7.16 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-15134.948, Time=7.03 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-16538.454, Time=8.41 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-16271.346, Time=2.22 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-16350.992, Time=12.88 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-16200.149, Time=3.37 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-16461.809, Time=15.31 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-16384.147, Time=3.25 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=10.13 sec
ARIMA(2,3,1)(0,0,0)[0] intercept : AIC=-15110.164, Time=5.64 sec
Best model: ARIMA(2,3,1)(0,0,0)[0]
Total fit time: 85.264 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(2, 3, 1) Log Likelihood 8294.227
Date: Sun, 12 Dec 2021 AIC -16538.454
Time: 14:05:38 BIC -16421.183
Sample: 0 HQIC -16493.417
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 3.591e-07 0.001 0.000 1.000 -0.002 0.002
x2 3.6e-07 0.002 0.000 1.000 -0.003 0.003
x3 3.611e-07 0.001 0.000 1.000 -0.002 0.002
x4 1.0000 0.000 2628.605 0.000 0.999 1.001
x5 3.432e-07 0.000 0.001 0.999 -0.001 0.001
x6 1.714e-07 4.05e-05 0.004 0.997 -7.91e-05 7.95e-05
x7 3.541e-07 0.001 0.000 1.000 -0.003 0.003
x8 -0.0002 0.000 -1.006 0.315 -0.001 0.000
x9 -7.559e-08 0.000 -0.000 1.000 -0.001 0.001
x10 0.0001 0.000 0.492 0.623 -0.000 0.001
x11 -0.0006 0.000 -2.697 0.007 -0.001 -0.000
x12 0.0005 0.000 1.741 0.082 -5.97e-05 0.001
x13 3.6e-07 0.000 0.002 0.999 -0.000 0.000
x14 1.003e-06 0.001 0.001 0.999 -0.002 0.002
x15 3.506e-07 7.16e-05 0.005 0.996 -0.000 0.000
x16 5.157e-07 0.000 0.005 0.996 -0.000 0.000
x17 3.516e-07 6.59e-05 0.005 0.996 -0.000 0.000
x18 1.166e-07 0.000 0.001 1.000 -0.000 0.000
x19 3.922e-07 7.5e-05 0.005 0.996 -0.000 0.000
x20 -3.64e-07 0.000 -0.002 0.999 -0.000 0.000
x21 4.458e-07 0.000 0.004 0.997 -0.000 0.000
ar.L1 -0.4019 4.12e-05 -9758.484 0.000 -0.402 -0.402
ar.L2 -0.1006 1.58e-05 -6360.873 0.000 -0.101 -0.101
ma.L1 -0.7963 8.45e-06 -9.43e+04 0.000 -0.796 -0.796
sigma2 9.048e-11 7.2e-11 1.257 0.209 -5.06e-11 2.32e-10
===================================================================================
Ljung-Box (L1) (Q): 64.02 Jarque-Bera (JB): 4424775.47
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.53
Prob(H) (two-sided): 0.00 Kurtosis: 366.04
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.02e+20. Standard errors may be unstable.
ARIMA order: (2, 3, 1)
WARNING:tensorflow:Layer lstm_24 will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.21133, saving model to LSTM5.h5
90/90 - 2s - loss: 0.1634 - val_loss: 0.2113 - lr: 0.0010 - 2s/epoch - 24ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.21133 to 0.04083, saving model to LSTM5.h5
90/90 - 1s - loss: 0.2226 - val_loss: 0.0408 - lr: 0.0010 - 666ms/epoch - 7ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.1013 - val_loss: 0.4993 - lr: 0.0010 - 670ms/epoch - 7ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0733 - val_loss: 0.0470 - lr: 0.0010 - 615ms/epoch - 7ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0373 - val_loss: 0.2935 - lr: 0.0010 - 659ms/epoch - 7ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0306 - val_loss: 0.1003 - lr: 0.0010 - 624ms/epoch - 7ms/step
Epoch 7/500
Epoch 00007: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00007: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0359 - val_loss: 0.0651 - lr: 0.0010 - 657ms/epoch - 7ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0318 - val_loss: 0.0661 - lr: 1.0000e-04 - 670ms/epoch - 7ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0313 - val_loss: 0.0655 - lr: 1.0000e-04 - 647ms/epoch - 7ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0285 - val_loss: 0.0612 - lr: 1.0000e-04 - 683ms/epoch - 8ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0303 - val_loss: 0.0535 - lr: 1.0000e-04 - 662ms/epoch - 7ms/step
Epoch 12/500
Epoch 00012: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00012: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0284 - val_loss: 0.0558 - lr: 1.0000e-04 - 625ms/epoch - 7ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0276 - val_loss: 0.0549 - lr: 1.0000e-05 - 697ms/epoch - 8ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0295 - val_loss: 0.0555 - lr: 1.0000e-05 - 661ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0274 - val_loss: 0.0549 - lr: 1.0000e-05 - 660ms/epoch - 7ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0283 - val_loss: 0.0536 - lr: 1.0000e-05 - 707ms/epoch - 8ms/step
Epoch 17/500
Epoch 00017: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00017: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0275 - val_loss: 0.0535 - lr: 1.0000e-05 - 649ms/epoch - 7ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0259 - val_loss: 0.0544 - lr: 1.0000e-05 - 631ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0279 - val_loss: 0.0537 - lr: 1.0000e-05 - 678ms/epoch - 8ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0259 - val_loss: 0.0540 - lr: 1.0000e-05 - 639ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0259 - val_loss: 0.0538 - lr: 1.0000e-05 - 622ms/epoch - 7ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0267 - val_loss: 0.0534 - lr: 1.0000e-05 - 655ms/epoch - 7ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0284 - val_loss: 0.0537 - lr: 1.0000e-05 - 662ms/epoch - 7ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0279 - val_loss: 0.0538 - lr: 1.0000e-05 - 617ms/epoch - 7ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0276 - val_loss: 0.0538 - lr: 1.0000e-05 - 704ms/epoch - 8ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0266 - val_loss: 0.0546 - lr: 1.0000e-05 - 653ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0267 - val_loss: 0.0544 - lr: 1.0000e-05 - 652ms/epoch - 7ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0257 - val_loss: 0.0547 - lr: 1.0000e-05 - 623ms/epoch - 7ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0275 - val_loss: 0.0548 - lr: 1.0000e-05 - 655ms/epoch - 7ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0275 - val_loss: 0.0541 - lr: 1.0000e-05 - 662ms/epoch - 7ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0295 - val_loss: 0.0531 - lr: 1.0000e-05 - 680ms/epoch - 8ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0251 - val_loss: 0.0529 - lr: 1.0000e-05 - 610ms/epoch - 7ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0247 - val_loss: 0.0529 - lr: 1.0000e-05 - 639ms/epoch - 7ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0252 - val_loss: 0.0523 - lr: 1.0000e-05 - 681ms/epoch - 8ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0274 - val_loss: 0.0533 - lr: 1.0000e-05 - 640ms/epoch - 7ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0256 - val_loss: 0.0519 - lr: 1.0000e-05 - 645ms/epoch - 7ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0287 - val_loss: 0.0513 - lr: 1.0000e-05 - 663ms/epoch - 7ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0264 - val_loss: 0.0515 - lr: 1.0000e-05 - 639ms/epoch - 7ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0270 - val_loss: 0.0501 - lr: 1.0000e-05 - 689ms/epoch - 8ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0253 - val_loss: 0.0489 - lr: 1.0000e-05 - 649ms/epoch - 7ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0279 - val_loss: 0.0484 - lr: 1.0000e-05 - 673ms/epoch - 7ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0268 - val_loss: 0.0494 - lr: 1.0000e-05 - 657ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0262 - val_loss: 0.0489 - lr: 1.0000e-05 - 664ms/epoch - 7ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0254 - val_loss: 0.0491 - lr: 1.0000e-05 - 617ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0237 - val_loss: 0.0488 - lr: 1.0000e-05 - 653ms/epoch - 7ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0257 - val_loss: 0.0473 - lr: 1.0000e-05 - 657ms/epoch - 7ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0258 - val_loss: 0.0473 - lr: 1.0000e-05 - 667ms/epoch - 7ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0269 - val_loss: 0.0478 - lr: 1.0000e-05 - 668ms/epoch - 7ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0271 - val_loss: 0.0471 - lr: 1.0000e-05 - 626ms/epoch - 7ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0270 - val_loss: 0.0481 - lr: 1.0000e-05 - 667ms/epoch - 7ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0242 - val_loss: 0.0491 - lr: 1.0000e-05 - 692ms/epoch - 8ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.04083
90/90 - 1s - loss: 0.0258 - val_loss: 0.0486 - lr: 1.0000e-05 - 665ms/epoch - 7ms/step
Epoch 00052: early stopping
SMA Prediction vs Close: 51.49% Accuracy Prediction vs Prediction: 48.13% Accuracy MSE: 38.984836670221576 RMSE: 6.24378384236847 MAPE: 5.10393861237253 EMA Prediction vs Close: 55.22% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 124.25764226707467 RMSE: 11.1470912020614 MAPE: 9.17724208981177 WMA Prediction vs Close: 54.48% Accuracy Prediction vs Prediction: 43.66% Accuracy MSE: 38.21405797999008 RMSE: 6.1817520154071275 MAPE: 5.0592557753421294 DEMA Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 42.54% Accuracy MSE: 272.80520892035037 RMSE: 16.51681594376926 MAPE: 15.690440427295842 KAMA Prediction vs Close: 55.6% Accuracy Prediction vs Prediction: 45.9% Accuracy MSE: 39.855043502438484 RMSE: 6.313085101789654 MAPE: 4.932118299391016 MIDPOINT Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 47.39% Accuracy MSE: 247.8059617737088 RMSE: 15.741853822650901 MAPE: 13.137429929578502 T3 Prediction vs Close: 54.85% Accuracy Prediction vs Prediction: 42.91% Accuracy MSE: 210.84512917819418 RMSE: 14.520507194247527 MAPE: 11.877711162377306 TEMA Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 47.76% Accuracy MSE: 47.81525777472662 RMSE: 6.9148577552055706 MAPE: 5.805355375153735 Runtime: mins: 43.70210928193333
Architecture Used¶
In [ ]:
from google.colab import files
import cv2
uploaded = files.upload()
Saving Experiment5.png to Experiment5 (1).png
In [ ]:
img = cv2.imread('Experiment5.png')
plt.figure(figsize=(20,10))
plt.axis("off")
plt.title('LSTM Architecture '+imgfile,fontsize=18)
plt.imshow(img)
Out[ ]:
<matplotlib.image.AxesImage at 0x7fda6cb8a950>
Model Plots¶
In [79]:
with open('simulation5_data.json') as json_file:
simulation5 = json.load(json_file)
fileimg = 'Experiment5'
In [80]:
for i in range(len(list(simulation5.keys()))):
SIM = list(simulation5.keys())[i]
plot_train(simulation5,SIM)
plot_test(simulation5,SIM)
----- Train RMSE for SMA ----- 7.825688029871665 ----- Train_MSE_LSTM for SMA ----- 61.24139314087665 ----- Train MAE LSTM for SMA ----- 6.696912122355534
----- Test RMSE for SMA----- 6.24378384236847 ----- Test_MSE_LSTM for SMA----- 38.984836670221576 ----- Test_MAE_LSTM for SMA----- 5.10393861237253
----- Train RMSE for EMA ----- 9.263860311359515 ----- Train_MSE_LSTM for EMA ----- 85.819107868382 ----- Train MAE LSTM for EMA ----- 8.131221693493242
----- Test RMSE for EMA----- 11.1470912020614 ----- Test_MSE_LSTM for EMA----- 124.25764226707467 ----- Test_MAE_LSTM for EMA----- 9.17724208981177
----- Train RMSE for WMA ----- 9.62296255691105 ----- Train_MSE_LSTM for WMA ----- 92.60140837171207 ----- Train MAE LSTM for WMA ----- 8.447782940991997
----- Test RMSE for WMA----- 6.1817520154071275 ----- Test_MSE_LSTM for WMA----- 38.21405797999008 ----- Test_MAE_LSTM for WMA----- 5.0592557753421294
----- Train RMSE for DEMA ----- 11.500185279159618 ----- Train_MSE_LSTM for DEMA ----- 132.25426145499958 ----- Train MAE LSTM for DEMA ----- 10.30295939891055
----- Test RMSE for DEMA----- 16.51681594376926 ----- Test_MSE_LSTM for DEMA----- 272.80520892035037 ----- Test_MAE_LSTM for DEMA----- 15.690440427295842
----- Train RMSE for KAMA ----- 9.40669400960287 ----- Train_MSE_LSTM for KAMA ----- 88.48589219029851 ----- Train MAE LSTM for KAMA ----- 8.417607692965897
----- Test RMSE for KAMA----- 6.313085101789654 ----- Test_MSE_LSTM for KAMA----- 39.855043502438484 ----- Test_MAE_LSTM for KAMA----- 4.932118299391016
----- Train RMSE for MIDPOINT ----- 8.65407178367174 ----- Train_MSE_LSTM for MIDPOINT ----- 74.89295843694336 ----- Train MAE LSTM for MIDPOINT ----- 7.619016227644343
----- Test RMSE for MIDPOINT----- 15.741853822650901 ----- Test_MSE_LSTM for MIDPOINT----- 247.8059617737088 ----- Test_MAE_LSTM for MIDPOINT----- 13.137429929578502
----- Train RMSE for T3 ----- 11.015384594469248 ----- Train_MSE_LSTM for T3 ----- 121.33869776407043 ----- Train MAE LSTM for T3 ----- 9.875190325929681
----- Test RMSE for T3----- 14.520507194247527 ----- Test_MSE_LSTM for T3----- 210.84512917819418 ----- Test_MAE_LSTM for T3----- 11.877711162377306
----- Train RMSE for TEMA ----- 6.884503600474202 ----- Train_MSE_LSTM for TEMA ----- 47.396389824942254 ----- Train MAE LSTM for TEMA ----- 4.709048845781707
----- Test RMSE for TEMA----- 6.9148577552055706 ----- Test_MSE_LSTM for TEMA----- 47.81525777472662 ----- Test_MAE_LSTM for TEMA----- 5.805355375153735
Arima w Exogenous Variable Multistep MutiVariate LSTM Hybrid Model Experiment 6¶
In [ ]:
def get_arima_exog(dataframe,original_data, train_len, test_len):
# prepare train and test data for exogenous vr
X_value = pd.DataFrame(low_vol.iloc[:, :])
y_value = pd.DataFrame(low_vol.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
X_scale_dataset = X_scaler.fit_transform(X_value)
y_scale_dataset = y_scaler.fit_transform(y_value)
# Get data and check shape
# X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X_scale_dataset)
y_train, y_test, = split_train_test(y_scale_dataset)
yc_train,yc_test = split_train_test(low_vol_data)
yc = yc_test.values.tolist()
y_train_list = y_train.flatten().tolist()
y_test_list = y_test.flatten().tolist()
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
# Initialize model
model = auto_arima(y_train_list,exogenous = X_train,trace=True, error_action='ignore', start_p=1,start_q=1,max_p=3,max_q=3,d=3,
suppress_warnings=True,stepwise=True,seasonal=True)
# Determine model parameters
print(model.summary())
model.fit(y_train_list,maxiter=200)
order = model.get_params()['order']
print('ARIMA order:', order, '\n')
# Genereate predictions
prediction = []
for i in range(len(y_test_list)):
model = pmdarima.ARIMA(order=order)
model.fit(y_train_list)
# print('working on', i+1, 'of', len(y_test), '-- ' + str(int(100 * (i + 1) / len(y_test))) + '% complete')
prediction.append(model.predict()[0])
y_train_list.append(y_test_list[i])
predictionte = y_scaler.inverse_transform(np.array(prediction).reshape(-1,1))
y_test_ = y_scaler.inverse_transform(np.array(y_test_list).reshape(-1,1))
# Generate error data
mse = mean_squared_error(yc_test, predictionte)
rmse = mse ** 0.5
mae = mean_absolute_error(y_test_ , predictionte )
return yc,predictionte.flatten().tolist(), mse, rmse, mae
In [ ]:
def get_lstm(data,original_data, train_len, test_len,img_file,ma ,lstm_len=3):
# prepare train and test data
X_value = pd.DataFrame(data.iloc[:, :])
y_value = pd.DataFrame(data.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
# Get data and check shape
X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X)
y_train, y_test, = split_train_test(y)
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
det =20
input_dim = X_train.shape[1]#3
feature_size = X_train.shape[2]#24
output_dim = y_train.shape[1]#1
# Option 1
# Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(256, activation='relu', kernel_initializer='he_normal', input_shape=(input_dim, feature_size)))
# model.add(Dense(units=64,activation='relu'))
# model.add(Dropout(0.5))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(learning_rate = 0.001), loss='mse')
# ## Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# # option 2
model = Sequential()
model.add(Bidirectional(LSTM(units= 128), input_shape=(input_dim, feature_size)))
model.add(Dense(64))
model.add(Dense(units=output_dim))
model.compile(optimizer=Adam(lr = 0.001), loss='mean_squared_error', metrics=['accuracy'])
# Common code
callbacks = [
EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
ModelCheckpoint('LSTM6.h5', verbose=1, save_best_only=True, save_weights_only=True)]
fname1 = img_file+'.png'
tensorflow.keras.utils.plot_model(
model, to_file=fname1, show_shapes=True, show_dtype=False,
show_layer_names=True, expand_nested=False, dpi=96,
layer_range=None, show_layer_activations=False
)
history = model.fit(X_train, y_train, epochs=500, batch_size=int( optimized_period[ma]), verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# plot loss
fname2 = img_file+'-'+ma
plt.title(img_file+'-'+ma+' Loss')
plt.xlabel("Epochs")
plt.ylabel("Loss")
pyplot.plot(history.history['loss'], label='train')
pyplot.plot(history.history['val_loss'], label='validation')
pyplot.legend()
pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# Option 3
# define custom activation
#
# class Double_Tanh(Activation):
# def __init__(self, activation, **kwargs):
# super(Double_Tanh, self).__init__(activation, **kwargs)
# self.__name__ = 'double_tanh'
# def double_tanh(x):
# return (K.tanh(x) * 2)
# get_custom_objects().update({'double_tanh':Double_Tanh(double_tanh)})
# # Model Generation
# model = Sequential()
# #check https://machinelearningmastery.com/use-weight-regularization-lstm-networks-time-series-forecasting/
# model.add(LSTM(25, input_shape=(input_dim, feature_size), dropout=0.2, kernel_regularizer=l1_l2(0.00,0.00), bias_regularizer=l1_l2(0.00,0.00)))
# model.add(Dense(1))
# model.add(Activation(double_tanh))
# model.compile(loss='mean_squared_error', optimizer='adam', metrics=['mse', 'mae'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Option 4
# Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(units=lstm_len, return_sequences=True, input_shape=(x_train.shape[1], 1)))
# model.add(LSTM(units=int(lstm_len/2)))
# model.add(Dense(1, activation='sigmoid'))
# model.compile(loss='mean_squared_error', optimizer='adam')
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Generate predictions
predictiontr = model.predict(X_train, verbose=0)
predictiontr = y_scaler.inverse_transform(predictiontr).tolist()
outputtr = []
for i in range(len(predictiontr)):
outputtr.extend(predictiontr[i])
predictiontr = outputtr
# Generate error data
## replace with yc , xtest generated by new multistep method
mse_tr = mean_squared_error(y_train, predictiontr)
rmse_tr = mse_tr ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictiontr))
mae_tr = mean_absolute_error(y_train, pd.Series(predictiontr))
# Original_tr = pd.Series(yc_train)
Original_tr = y_scaler.inverse_transform(y_train).flatten().tolist()
predictionte = model.predict(X_test, verbose=0)
predictionte = (y_scaler.inverse_transform(predictionte)-det).tolist()
outputte = []
for i in range(len(predictionte)):
outputte.extend(predictionte[i])
predictionte = outputte
# Generate error data
mse_te = mean_squared_error(y_test, predictionte)
rmse_te = mse_te ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictionte))
mae_te = mean_absolute_error(y_test, pd.Series(predictionte))
# Original_te = pd.Series(yc_test)
Original_te = y_scaler.inverse_transform(y_test).flatten().tolist()
return Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,Original_te,predictionte, mse_te,rmse_te,mae_te
In [ ]:
if __name__ == '__main__':
start_time = timeit.default_timer()
simulation6 = {}
imgfile = 'Experiment6'
for ma in optimized_period:
print(ma)
print(functions[ma])
print ( int( optimized_period[ma]))
# if ma == 'SMA':
low_vol = df.apply(lambda c: functions[ma](c, timeperiod = int( optimized_period[ma])))
low_vol = low_vol.fillna(0)
low_vol_data = df['close']
high_vol = pd.DataFrame()
df2 = df.copy()
for i in df2.columns:
if i in low_vol.columns:
high_vol[i] = df2[i].subtract(low_vol[i], fill_value=0)
high_vol_data = df['close']
## *****************************************************
# Generate ARIMA and LSTM predictions
print('\nWorking on ' + ma + ' predictions')
try:
print('parameters used : ', train_len, test_len)
low_vol_Original, low_vol_prediction, low_vol_mse, low_vol_rmse,low_vol_mae = get_arima_exog(low_vol,low_vol_data, train_len, test_len)
except:
print('ARIMA error, skipping to next MA type')
continue
Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,high_vol_Original, high_vol_prediction, high_vol_mse, high_vol_rmse,high_vol_mae, = get_lstm(high_vol,high_vol_data, train_len, test_len,imgfile,ma)
final_prediction_tr = df['close'].head(train_len).values + pd.Series(predictiontr) # ignoring first 3 steps
mse_ftr = mean_squared_error(df['close'].head(train_len).values,final_prediction_tr.values)
rmse_ftr = mse_ftr ** 0.5
mape_ftr = mean_absolute_percentage_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
mae_ftr = mean_absolute_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
final_prediction = pd.Series(low_vol_prediction[3:]) + pd.Series(high_vol_prediction)
mse = mean_squared_error(df['close'].tail(test_len).values,final_prediction.values)
rmse = mse ** 0.5
mape = mean_absolute_percentage_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
mae = mean_absolute_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
# Generate prediction accuracy
actual = df['close'].tail(test_len).values
result_1 = []
result_2 = []
for i in range(1, len(final_prediction)):
# Compare prediction to previous close price
if final_prediction[i] > actual[i-1] and actual[i] > actual[i-1]:
result_1.append(1)
elif final_prediction[i] < actual[i-1] and actual[i] < actual[i-1]:
result_1.append(1)
else:
result_1.append(0)
# Compare prediction to previous prediction
if final_prediction[i] > final_prediction[i-1] and actual[i] > actual[i-1]:
result_2.append(1)
elif final_prediction[i] < final_prediction[i-1] and actual[i] < actual[i-1]:
result_2.append(1)
else:
result_2.append(0)
accuracy_1 = np.mean(result_1)
accuracy_2 = np.mean(result_2)
simulation6[ma] = {'low_vol': {'original':list(low_vol_Original), 'prediction': list(low_vol_prediction) , 'mse': low_vol_mse,
'rmse': low_vol_rmse, 'mae' : low_vol_mae},
'high_vol': {'original':list(high_vol_Original),'prediction': list(high_vol_prediction), 'mse': high_vol_mse,
'rmse': high_vol_rmse, 'mae' : high_vol_mae},
'final_tr': {'original':df['close'].head(train_len).tolist(),'prediction': final_prediction_tr.values.tolist(), 'mse': mse_ftr,
'rmse': rmse_ftr, 'mae' : mae_ftr},
'final': {'original': df['close'].tail(test_len).tolist(), 'prediction': final_prediction.values.tolist(), 'mse': mse,
'rmse': rmse, 'mae': mae },
'accuracy': {'prediction vs close': accuracy_1, 'prediction vs prediction': accuracy_2}}
# save simulation data here as checkpoint
with open('simulation6_data.json', 'w') as fp:
json.dump(simulation6, fp)
for ma in simulation6.keys():
print('\n' + ma)
print('Prediction vs Close:\t\t' + str(round(100*simulation6[ma]['accuracy']['prediction vs close'], 2))
+ '% Accuracy')
print('Prediction vs Prediction:\t' + str(round(100*simulation6[ma]['accuracy']['prediction vs prediction'], 2))
+ '% Accuracy')
print('MSE:\t', simulation6[ma]['final']['mse'],
'\nRMSE:\t', simulation6[ma]['final']['rmse'],
'\nMAPE:\t', simulation6[ma]['final']['mae'])#,
# '\nMAPE:\t', simulation[ma]['final']['mape'])
# else:
# break
elapsed = timeit.default_timer() - start_time
print('Runtime: mins:',elapsed/60)
SMA
SMA([input_arrays], [timeperiod=30])
Simple Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
17
Working on SMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-15000.708, Time=8.47 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-13492.284, Time=2.32 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15827.971, Time=7.98 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-13635.197, Time=9.87 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-14132.778, Time=3.91 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15140.312, Time=9.72 sec
ARIMA(1,3,0)(0,0,0)[0] intercept : AIC=-13970.469, Time=7.17 sec
Best model: ARIMA(1,3,0)(0,0,0)[0]
Total fit time: 49.461 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(1, 3, 0) Log Likelihood 7936.985
Date: Sun, 12 Dec 2021 AIC -15827.971
Time: 16:37:24 BIC -15720.081
Sample: 0 HQIC -15786.537
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -4.786e-05 0.001 -0.066 0.947 -0.001 0.001
x2 -4.789e-05 0.001 -0.085 0.932 -0.001 0.001
x3 -4.819e-05 0.000 -0.105 0.917 -0.001 0.001
x4 1.0000 0.001 1557.248 0.000 0.999 1.001
x5 -4.579e-05 0.001 -0.071 0.943 -0.001 0.001
x6 -5.16e-05 0.000 -0.432 0.666 -0.000 0.000
x7 -4.778e-05 0.000 -0.278 0.781 -0.000 0.000
x8 -0.0012 0.000 -7.403 0.000 -0.002 -0.001
x9 -3.454e-06 0.002 -0.002 0.998 -0.003 0.003
x10 -0.0005 0.001 -0.403 0.687 -0.003 0.002
x11 0.0029 0.000 10.904 0.000 0.002 0.003
x12 -0.0003 0.000 -1.815 0.069 -0.001 2.06e-05
x13 -4.809e-05 0.000 -0.157 0.875 -0.001 0.001
x14 -0.0001 0.000 -0.482 0.630 -0.001 0.000
x15 -5.214e-05 0.000 -0.273 0.785 -0.000 0.000
x16 -4.468e-05 0.000 -0.125 0.901 -0.001 0.001
x17 -4.224e-05 0.000 -0.202 0.840 -0.000 0.000
x18 -8.086e-05 0.000 -0.270 0.787 -0.001 0.001
x19 -5.537e-05 0.000 -0.244 0.807 -0.000 0.000
x20 8.423e-05 0.000 0.333 0.739 -0.000 0.001
x21 -4.232e-05 0.000 -0.166 0.868 -0.001 0.000
ar.L1 -0.6666 6.03e-06 -1.11e+05 0.000 -0.667 -0.667
sigma2 4.093e-10 8.97e-11 4.563 0.000 2.33e-10 5.85e-10
===================================================================================
Ljung-Box (L1) (Q): 60.24 Jarque-Bera (JB): 1334882.31
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.11 Skew: -3.81
Prob(H) (two-sided): 0.00 Kurtosis: 202.35
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 5.73e+20. Standard errors may be unstable.
ARIMA order: (1, 3, 0)
/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead. super(Adam, self).__init__(name, **kwargs)
Epoch 1/500 Epoch 00001: val_loss improved from inf to 0.07591, saving model to LSTM6.h5 48/48 - 4s - loss: 0.1164 - accuracy: 0.0000e+00 - val_loss: 0.0759 - val_accuracy: 0.0037 - lr: 0.0010 - 4s/epoch - 81ms/step Epoch 2/500 Epoch 00002: val_loss improved from 0.07591 to 0.01860, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0536 - accuracy: 0.0000e+00 - val_loss: 0.0186 - val_accuracy: 0.0037 - lr: 0.0010 - 246ms/epoch - 5ms/step Epoch 3/500 Epoch 00003: val_loss did not improve from 0.01860 48/48 - 0s - loss: 0.0141 - accuracy: 0.0000e+00 - val_loss: 0.0443 - val_accuracy: 0.0037 - lr: 0.0010 - 230ms/epoch - 5ms/step Epoch 4/500 Epoch 00004: val_loss improved from 0.01860 to 0.00820, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0156 - accuracy: 0.0000e+00 - val_loss: 0.0082 - val_accuracy: 0.0037 - lr: 0.0010 - 263ms/epoch - 5ms/step Epoch 5/500 Epoch 00005: val_loss did not improve from 0.00820 48/48 - 0s - loss: 0.0028 - accuracy: 0.0000e+00 - val_loss: 0.0182 - val_accuracy: 0.0037 - lr: 0.0010 - 235ms/epoch - 5ms/step Epoch 6/500 Epoch 00006: val_loss improved from 0.00820 to 0.00325, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0035 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 0.0010 - 274ms/epoch - 6ms/step Epoch 7/500 Epoch 00007: val_loss did not improve from 0.00325 48/48 - 0s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0102 - val_accuracy: 0.0037 - lr: 0.0010 - 220ms/epoch - 5ms/step Epoch 8/500 Epoch 00008: val_loss did not improve from 0.00325 48/48 - 0s - loss: 0.0019 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 0.0010 - 232ms/epoch - 5ms/step Epoch 9/500 Epoch 00009: val_loss did not improve from 0.00325 48/48 - 0s - loss: 0.0019 - accuracy: 0.0000e+00 - val_loss: 0.0088 - val_accuracy: 0.0037 - lr: 0.0010 - 238ms/epoch - 5ms/step Epoch 10/500 Epoch 00010: val_loss did not improve from 0.00325 48/48 - 0s - loss: 0.0043 - accuracy: 0.0000e+00 - val_loss: 0.0061 - val_accuracy: 0.0037 - lr: 0.0010 - 223ms/epoch - 5ms/step Epoch 11/500 Epoch 00011: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513. Epoch 00011: val_loss did not improve from 0.00325 48/48 - 0s - loss: 0.0119 - accuracy: 0.0000e+00 - val_loss: 0.0064 - val_accuracy: 0.0037 - lr: 0.0010 - 218ms/epoch - 5ms/step Epoch 12/500 Epoch 00012: val_loss did not improve from 0.00325 48/48 - 0s - loss: 0.0213 - accuracy: 0.0000e+00 - val_loss: 0.0077 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 244ms/epoch - 5ms/step Epoch 13/500 Epoch 00013: val_loss did not improve from 0.00325 48/48 - 0s - loss: 0.0040 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 260ms/epoch - 5ms/step Epoch 14/500 Epoch 00014: val_loss did not improve from 0.00325 48/48 - 0s - loss: 0.0028 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 229ms/epoch - 5ms/step Epoch 15/500 Epoch 00015: val_loss did not improve from 0.00325 48/48 - 0s - loss: 0.0019 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 226ms/epoch - 5ms/step Epoch 16/500 Epoch 00016: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05. Epoch 00016: val_loss improved from 0.00325 to 0.00320, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 241ms/epoch - 5ms/step Epoch 17/500 Epoch 00017: val_loss improved from 0.00320 to 0.00320, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 287ms/epoch - 6ms/step Epoch 18/500 Epoch 00018: val_loss improved from 0.00320 to 0.00319, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 248ms/epoch - 5ms/step Epoch 19/500 Epoch 00019: val_loss improved from 0.00319 to 0.00317, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 237ms/epoch - 5ms/step Epoch 20/500 Epoch 00020: val_loss improved from 0.00317 to 0.00315, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 243ms/epoch - 5ms/step Epoch 21/500 Epoch 00021: val_loss improved from 0.00315 to 0.00313, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 280ms/epoch - 6ms/step Epoch 22/500 Epoch 00022: val_loss improved from 0.00313 to 0.00310, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 244ms/epoch - 5ms/step Epoch 23/500 Epoch 00023: val_loss improved from 0.00310 to 0.00308, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 243ms/epoch - 5ms/step Epoch 24/500 Epoch 00024: val_loss improved from 0.00308 to 0.00306, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 247ms/epoch - 5ms/step Epoch 25/500 Epoch 00025: val_loss improved from 0.00306 to 0.00304, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 264ms/epoch - 6ms/step Epoch 26/500 Epoch 00026: val_loss improved from 0.00304 to 0.00302, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 243ms/epoch - 5ms/step Epoch 27/500 Epoch 00027: val_loss improved from 0.00302 to 0.00300, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 254ms/epoch - 5ms/step Epoch 28/500 Epoch 00028: val_loss improved from 0.00300 to 0.00299, saving model to LSTM6.h5 48/48 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 267ms/epoch - 6ms/step Epoch 29/500 Epoch 00029: val_loss improved from 0.00299 to 0.00297, saving model to LSTM6.h5 48/48 - 0s - loss: 9.9052e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 250ms/epoch - 5ms/step Epoch 30/500 Epoch 00030: ReduceLROnPlateau reducing learning rate to 1e-05. Epoch 00030: val_loss improved from 0.00297 to 0.00296, saving model to LSTM6.h5 48/48 - 0s - loss: 9.8140e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 255ms/epoch - 5ms/step Epoch 31/500 Epoch 00031: val_loss improved from 0.00296 to 0.00295, saving model to LSTM6.h5 48/48 - 0s - loss: 9.7271e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 242ms/epoch - 5ms/step Epoch 32/500 Epoch 00032: val_loss improved from 0.00295 to 0.00294, saving model to LSTM6.h5 48/48 - 0s - loss: 9.6446e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 251ms/epoch - 5ms/step Epoch 33/500 Epoch 00033: val_loss improved from 0.00294 to 0.00293, saving model to LSTM6.h5 48/48 - 0s - loss: 9.5663e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 292ms/epoch - 6ms/step Epoch 34/500 Epoch 00034: val_loss improved from 0.00293 to 0.00292, saving model to LSTM6.h5 48/48 - 0s - loss: 9.4922e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 243ms/epoch - 5ms/step Epoch 35/500 Epoch 00035: val_loss improved from 0.00292 to 0.00292, saving model to LSTM6.h5 48/48 - 0s - loss: 9.4222e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 239ms/epoch - 5ms/step Epoch 36/500 Epoch 00036: val_loss improved from 0.00292 to 0.00291, saving model to LSTM6.h5 48/48 - 0s - loss: 9.3562e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 259ms/epoch - 5ms/step Epoch 37/500 Epoch 00037: val_loss improved from 0.00291 to 0.00291, saving model to LSTM6.h5 48/48 - 0s - loss: 9.2941e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 258ms/epoch - 5ms/step Epoch 38/500 Epoch 00038: val_loss did not improve from 0.00291 48/48 - 0s - loss: 9.2358e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step Epoch 39/500 Epoch 00039: val_loss did not improve from 0.00291 48/48 - 0s - loss: 9.1810e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 231ms/epoch - 5ms/step Epoch 40/500 Epoch 00040: val_loss did not improve from 0.00291 48/48 - 0s - loss: 9.1297e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 237ms/epoch - 5ms/step Epoch 41/500 Epoch 00041: val_loss did not improve from 0.00291 48/48 - 0s - loss: 9.0817e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 234ms/epoch - 5ms/step Epoch 42/500 Epoch 00042: val_loss did not improve from 0.00291 48/48 - 0s - loss: 9.0367e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 232ms/epoch - 5ms/step Epoch 43/500 Epoch 00043: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.9947e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step Epoch 44/500 Epoch 00044: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.9553e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 249ms/epoch - 5ms/step Epoch 45/500 Epoch 00045: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.9186e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 257ms/epoch - 5ms/step Epoch 46/500 Epoch 00046: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.8842e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 235ms/epoch - 5ms/step Epoch 47/500 Epoch 00047: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.8520e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step Epoch 48/500 Epoch 00048: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.8218e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 231ms/epoch - 5ms/step Epoch 49/500 Epoch 00049: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.7934e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 261ms/epoch - 5ms/step Epoch 50/500 Epoch 00050: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.7667e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step Epoch 51/500 Epoch 00051: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.7416e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 234ms/epoch - 5ms/step Epoch 52/500 Epoch 00052: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.7178e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 233ms/epoch - 5ms/step Epoch 53/500 Epoch 00053: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.6953e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 242ms/epoch - 5ms/step Epoch 54/500 Epoch 00054: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.6738e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 235ms/epoch - 5ms/step Epoch 55/500 Epoch 00055: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.6533e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step Epoch 56/500 Epoch 00056: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.6337e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step Epoch 57/500 Epoch 00057: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.6149e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 241ms/epoch - 5ms/step Epoch 58/500 Epoch 00058: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.5966e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 227ms/epoch - 5ms/step Epoch 59/500 Epoch 00059: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.5790e-04 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step Epoch 60/500 Epoch 00060: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.5618e-04 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 232ms/epoch - 5ms/step Epoch 61/500 Epoch 00061: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.5450e-04 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 239ms/epoch - 5ms/step Epoch 62/500 Epoch 00062: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.5286e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step Epoch 63/500 Epoch 00063: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.5124e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 229ms/epoch - 5ms/step Epoch 64/500 Epoch 00064: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.4964e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 249ms/epoch - 5ms/step Epoch 65/500 Epoch 00065: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.4807e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 237ms/epoch - 5ms/step Epoch 66/500 Epoch 00066: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.4650e-04 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 233ms/epoch - 5ms/step Epoch 67/500 Epoch 00067: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.4494e-04 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 232ms/epoch - 5ms/step Epoch 68/500 Epoch 00068: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.4339e-04 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 235ms/epoch - 5ms/step Epoch 69/500 Epoch 00069: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.4184e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step Epoch 70/500 Epoch 00070: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.4028e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 241ms/epoch - 5ms/step Epoch 71/500 Epoch 00071: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.3873e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step Epoch 72/500 Epoch 00072: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.3717e-04 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step Epoch 73/500 Epoch 00073: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.3561e-04 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 230ms/epoch - 5ms/step Epoch 74/500 Epoch 00074: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.3404e-04 - accuracy: 0.0000e+00 - val_loss: 0.0038 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step Epoch 75/500 Epoch 00075: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.3246e-04 - accuracy: 0.0000e+00 - val_loss: 0.0038 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step Epoch 76/500 Epoch 00076: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.3087e-04 - accuracy: 0.0000e+00 - val_loss: 0.0038 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 212ms/epoch - 4ms/step Epoch 77/500 Epoch 00077: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.2927e-04 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step Epoch 78/500 Epoch 00078: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.2766e-04 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 227ms/epoch - 5ms/step Epoch 79/500 Epoch 00079: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.2604e-04 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step Epoch 80/500 Epoch 00080: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.2441e-04 - accuracy: 0.0000e+00 - val_loss: 0.0040 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step Epoch 81/500 Epoch 00081: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.2276e-04 - accuracy: 0.0000e+00 - val_loss: 0.0040 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 243ms/epoch - 5ms/step Epoch 82/500 Epoch 00082: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.2110e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step Epoch 83/500 Epoch 00083: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.1943e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 232ms/epoch - 5ms/step Epoch 84/500 Epoch 00084: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.1775e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 231ms/epoch - 5ms/step Epoch 85/500 Epoch 00085: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.1606e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 234ms/epoch - 5ms/step Epoch 86/500 Epoch 00086: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.1435e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 245ms/epoch - 5ms/step Epoch 87/500 Epoch 00087: val_loss did not improve from 0.00291 48/48 - 0s - loss: 8.1263e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step Epoch 00087: early stopping
SMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 71.66891031373025
RMSE: 8.465749247038342
MAPE: 6.880610177712922
EMA
EMA([input_arrays], [timeperiod=30])
Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
51
Working on EMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.775, Time=2.24 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=3.95 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16801.081, Time=8.73 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.16 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-14532.068, Time=6.53 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15610.472, Time=11.59 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16103.302, Time=13.63 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17030.021, Time=4.20 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.614, Time=3.13 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.441, Time=5.81 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=15.11 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17003.984, Time=3.06 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-16998.666, Time=3.42 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 86.580 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.720
Date: Sun, 12 Dec 2021 AIC -17087.441
Time: 16:39:56 BIC -16965.479
Sample: 0 HQIC -17040.602
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.316e-10 9.87e-21 -2.35e+10 0.000 -2.32e-10 -2.32e-10
x2 -2.308e-10 9.85e-21 -2.34e+10 0.000 -2.31e-10 -2.31e-10
x3 -2.324e-10 9.88e-21 -2.35e+10 0.000 -2.32e-10 -2.32e-10
x4 1.0000 9.87e-21 1.01e+20 0.000 1.000 1.000
x5 -2.106e-10 9.41e-21 -2.24e+10 0.000 -2.11e-10 -2.11e-10
x6 -7.996e-10 1.74e-20 -4.59e+10 0.000 -8e-10 -8e-10
x7 -2.295e-10 9.82e-21 -2.34e+10 0.000 -2.29e-10 -2.29e-10
x8 -2.244e-10 9.71e-21 -2.31e+10 0.000 -2.24e-10 -2.24e-10
x9 -1.166e-11 1.98e-21 -5.9e+09 0.000 -1.17e-11 -1.17e-11
x10 -4.453e-11 4.22e-21 -1.06e+10 0.000 -4.45e-11 -4.45e-11
x11 -2.219e-10 9.65e-21 -2.3e+10 0.000 -2.22e-10 -2.22e-10
x12 -2.264e-10 9.76e-21 -2.32e+10 0.000 -2.26e-10 -2.26e-10
x13 -2.315e-10 9.87e-21 -2.35e+10 0.000 -2.31e-10 -2.31e-10
x14 -1.766e-09 2.73e-20 -6.48e+10 0.000 -1.77e-09 -1.77e-09
x15 -2.167e-10 9.37e-21 -2.31e+10 0.000 -2.17e-10 -2.17e-10
x16 -5.232e-10 1.49e-20 -3.52e+10 0.000 -5.23e-10 -5.23e-10
x17 -2.147e-10 9.48e-21 -2.27e+10 0.000 -2.15e-10 -2.15e-10
x18 -3.791e-11 3.96e-21 -9.56e+09 0.000 -3.79e-11 -3.79e-11
x19 -2.597e-10 1.05e-20 -2.48e+10 0.000 -2.6e-10 -2.6e-10
x20 -2.417e-10 1e-20 -2.41e+10 0.000 -2.42e-10 -2.42e-10
x21 -4.823e-10 1.4e-20 -3.44e+10 0.000 -4.82e-10 -4.82e-10
ar.L1 -0.4923 1.46e-22 -3.38e+21 0.000 -0.492 -0.492
ar.L2 -0.1923 8.47e-23 -2.27e+21 0.000 -0.192 -0.192
ar.L3 -0.0462 4.02e-23 -1.15e+21 0.000 -0.046 -0.046
ma.L1 -0.7077 3.31e-22 -2.14e+21 0.000 -0.708 -0.708
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 54.09 Jarque-Bera (JB): 4207353.17
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.48
Prob(H) (two-sided): 0.00 Kurtosis: 357.00
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 3.15e+40. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead. super(Adam, self).__init__(name, **kwargs)
Epoch 1/500 Epoch 00001: val_loss improved from inf to 0.01113, saving model to LSTM6.h5 16/16 - 3s - loss: 0.0891 - accuracy: 0.0000e+00 - val_loss: 0.0111 - val_accuracy: 0.0037 - lr: 0.0010 - 3s/epoch - 216ms/step Epoch 2/500 Epoch 00002: val_loss did not improve from 0.01113 16/16 - 0s - loss: 0.0226 - accuracy: 0.0000e+00 - val_loss: 0.0154 - val_accuracy: 0.0037 - lr: 0.0010 - 97ms/epoch - 6ms/step Epoch 3/500 Epoch 00003: val_loss did not improve from 0.01113 16/16 - 0s - loss: 0.0111 - accuracy: 0.0000e+00 - val_loss: 0.0118 - val_accuracy: 0.0037 - lr: 0.0010 - 89ms/epoch - 6ms/step Epoch 4/500 Epoch 00004: val_loss did not improve from 0.01113 16/16 - 0s - loss: 0.0026 - accuracy: 0.0000e+00 - val_loss: 0.0246 - val_accuracy: 0.0037 - lr: 0.0010 - 98ms/epoch - 6ms/step Epoch 5/500 Epoch 00005: val_loss did not improve from 0.01113 16/16 - 0s - loss: 0.0023 - accuracy: 0.0000e+00 - val_loss: 0.0205 - val_accuracy: 0.0037 - lr: 0.0010 - 89ms/epoch - 6ms/step Epoch 6/500 Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513. Epoch 00006: val_loss did not improve from 0.01113 16/16 - 0s - loss: 0.0032 - accuracy: 0.0000e+00 - val_loss: 0.0169 - val_accuracy: 0.0037 - lr: 0.0010 - 92ms/epoch - 6ms/step Epoch 7/500 Epoch 00007: val_loss improved from 0.01113 to 0.01076, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0040 - accuracy: 0.0000e+00 - val_loss: 0.0108 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 104ms/epoch - 7ms/step Epoch 8/500 Epoch 00008: val_loss improved from 0.01076 to 0.01024, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0102 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 129ms/epoch - 8ms/step Epoch 9/500 Epoch 00009: val_loss did not improve from 0.01024 16/16 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0108 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 90ms/epoch - 6ms/step Epoch 10/500 Epoch 00010: val_loss improved from 0.01024 to 0.01020, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0102 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 112ms/epoch - 7ms/step Epoch 11/500 Epoch 00011: val_loss improved from 0.01020 to 0.00984, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0098 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 115ms/epoch - 7ms/step Epoch 12/500 Epoch 00012: val_loss improved from 0.00984 to 0.00959, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0096 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 110ms/epoch - 7ms/step Epoch 13/500 Epoch 00013: val_loss improved from 0.00959 to 0.00926, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0093 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 103ms/epoch - 6ms/step Epoch 14/500 Epoch 00014: val_loss improved from 0.00926 to 0.00896, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0090 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 107ms/epoch - 7ms/step Epoch 15/500 Epoch 00015: val_loss improved from 0.00896 to 0.00867, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0087 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 109ms/epoch - 7ms/step Epoch 16/500 Epoch 00016: val_loss improved from 0.00867 to 0.00840, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0084 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 105ms/epoch - 7ms/step Epoch 17/500 Epoch 00017: val_loss improved from 0.00840 to 0.00815, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0081 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 109ms/epoch - 7ms/step Epoch 18/500 Epoch 00018: val_loss improved from 0.00815 to 0.00790, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0079 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 113ms/epoch - 7ms/step Epoch 19/500 Epoch 00019: val_loss improved from 0.00790 to 0.00768, saving model to LSTM6.h5 16/16 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0077 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 102ms/epoch - 6ms/step Epoch 20/500 Epoch 00020: val_loss improved from 0.00768 to 0.00747, saving model to LSTM6.h5 16/16 - 0s - loss: 9.9982e-04 - accuracy: 0.0000e+00 - val_loss: 0.0075 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 117ms/epoch - 7ms/step Epoch 21/500 Epoch 00021: val_loss improved from 0.00747 to 0.00727, saving model to LSTM6.h5 16/16 - 0s - loss: 9.9141e-04 - accuracy: 0.0000e+00 - val_loss: 0.0073 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 115ms/epoch - 7ms/step Epoch 22/500 Epoch 00022: val_loss improved from 0.00727 to 0.00708, saving model to LSTM6.h5 16/16 - 0s - loss: 9.8312e-04 - accuracy: 0.0000e+00 - val_loss: 0.0071 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 111ms/epoch - 7ms/step Epoch 23/500 Epoch 00023: val_loss improved from 0.00708 to 0.00691, saving model to LSTM6.h5 16/16 - 0s - loss: 9.7497e-04 - accuracy: 0.0000e+00 - val_loss: 0.0069 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 109ms/epoch - 7ms/step Epoch 24/500 Epoch 00024: val_loss improved from 0.00691 to 0.00675, saving model to LSTM6.h5 16/16 - 0s - loss: 9.6695e-04 - accuracy: 0.0000e+00 - val_loss: 0.0067 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 110ms/epoch - 7ms/step Epoch 25/500 Epoch 00025: val_loss improved from 0.00675 to 0.00660, saving model to LSTM6.h5 16/16 - 0s - loss: 9.5909e-04 - accuracy: 0.0000e+00 - val_loss: 0.0066 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 107ms/epoch - 7ms/step Epoch 26/500 Epoch 00026: val_loss improved from 0.00660 to 0.00646, saving model to LSTM6.h5 16/16 - 0s - loss: 9.5138e-04 - accuracy: 0.0000e+00 - val_loss: 0.0065 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 111ms/epoch - 7ms/step Epoch 27/500 Epoch 00027: val_loss improved from 0.00646 to 0.00633, saving model to LSTM6.h5 16/16 - 0s - loss: 9.4382e-04 - accuracy: 0.0000e+00 - val_loss: 0.0063 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 172ms/epoch - 11ms/step Epoch 28/500 Epoch 00028: val_loss improved from 0.00633 to 0.00621, saving model to LSTM6.h5 16/16 - 0s - loss: 9.3643e-04 - accuracy: 0.0000e+00 - val_loss: 0.0062 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 107ms/epoch - 7ms/step Epoch 29/500 Epoch 00029: val_loss improved from 0.00621 to 0.00610, saving model to LSTM6.h5 16/16 - 0s - loss: 9.2920e-04 - accuracy: 0.0000e+00 - val_loss: 0.0061 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 111ms/epoch - 7ms/step Epoch 30/500 Epoch 00030: val_loss improved from 0.00610 to 0.00600, saving model to LSTM6.h5 16/16 - 0s - loss: 9.2214e-04 - accuracy: 0.0000e+00 - val_loss: 0.0060 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 108ms/epoch - 7ms/step Epoch 31/500 Epoch 00031: val_loss improved from 0.00600 to 0.00591, saving model to LSTM6.h5 16/16 - 0s - loss: 9.1523e-04 - accuracy: 0.0000e+00 - val_loss: 0.0059 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 109ms/epoch - 7ms/step Epoch 32/500 Epoch 00032: val_loss improved from 0.00591 to 0.00582, saving model to LSTM6.h5 16/16 - 0s - loss: 9.0850e-04 - accuracy: 0.0000e+00 - val_loss: 0.0058 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 103ms/epoch - 6ms/step Epoch 33/500 Epoch 00033: val_loss improved from 0.00582 to 0.00575, saving model to LSTM6.h5 16/16 - 0s - loss: 9.0192e-04 - accuracy: 0.0000e+00 - val_loss: 0.0058 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 110ms/epoch - 7ms/step Epoch 34/500 Epoch 00034: val_loss improved from 0.00575 to 0.00569, saving model to LSTM6.h5 16/16 - 0s - loss: 8.9551e-04 - accuracy: 0.0000e+00 - val_loss: 0.0057 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 112ms/epoch - 7ms/step Epoch 35/500 Epoch 00035: val_loss improved from 0.00569 to 0.00563, saving model to LSTM6.h5 16/16 - 0s - loss: 8.8926e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 110ms/epoch - 7ms/step Epoch 36/500 Epoch 00036: val_loss improved from 0.00563 to 0.00558, saving model to LSTM6.h5 16/16 - 0s - loss: 8.8317e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 109ms/epoch - 7ms/step Epoch 37/500 Epoch 00037: val_loss improved from 0.00558 to 0.00554, saving model to LSTM6.h5 16/16 - 0s - loss: 8.7724e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 118ms/epoch - 7ms/step Epoch 38/500 Epoch 00038: val_loss improved from 0.00554 to 0.00551, saving model to LSTM6.h5 16/16 - 0s - loss: 8.7146e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 113ms/epoch - 7ms/step Epoch 39/500 Epoch 00039: val_loss improved from 0.00551 to 0.00548, saving model to LSTM6.h5 16/16 - 0s - loss: 8.6583e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 106ms/epoch - 7ms/step Epoch 40/500 Epoch 00040: val_loss improved from 0.00548 to 0.00546, saving model to LSTM6.h5 16/16 - 0s - loss: 8.6034e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 107ms/epoch - 7ms/step Epoch 41/500 Epoch 00041: val_loss improved from 0.00546 to 0.00544, saving model to LSTM6.h5 16/16 - 0s - loss: 8.5501e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 105ms/epoch - 7ms/step Epoch 42/500 Epoch 00042: val_loss improved from 0.00544 to 0.00544, saving model to LSTM6.h5 16/16 - 0s - loss: 8.4981e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 104ms/epoch - 6ms/step Epoch 43/500 Epoch 00043: val_loss improved from 0.00544 to 0.00543, saving model to LSTM6.h5 16/16 - 0s - loss: 8.4476e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 109ms/epoch - 7ms/step Epoch 44/500 Epoch 00044: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05. Epoch 00044: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.3984e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 96ms/epoch - 6ms/step Epoch 45/500 Epoch 00045: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.3053e-04 - accuracy: 0.0000e+00 - val_loss: 0.0054 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step Epoch 46/500 Epoch 00046: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2878e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 88ms/epoch - 5ms/step Epoch 47/500 Epoch 00047: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2769e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 48/500 Epoch 00048: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2704e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step Epoch 49/500 Epoch 00049: ReduceLROnPlateau reducing learning rate to 1e-05. Epoch 00049: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2650e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step Epoch 50/500 Epoch 00050: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2600e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step Epoch 51/500 Epoch 00051: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2552e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step Epoch 52/500 Epoch 00052: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2502e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 92ms/epoch - 6ms/step Epoch 53/500 Epoch 00053: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2452e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 92ms/epoch - 6ms/step Epoch 54/500 Epoch 00054: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2402e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 88ms/epoch - 5ms/step Epoch 55/500 Epoch 00055: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2351e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step Epoch 56/500 Epoch 00056: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2299e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step Epoch 57/500 Epoch 00057: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2247e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 58/500 Epoch 00058: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2195e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step Epoch 59/500 Epoch 00059: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2142e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step Epoch 60/500 Epoch 00060: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2089e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step Epoch 61/500 Epoch 00061: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.2035e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step Epoch 62/500 Epoch 00062: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1981e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 92ms/epoch - 6ms/step Epoch 63/500 Epoch 00063: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1926e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step Epoch 64/500 Epoch 00064: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1871e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step Epoch 65/500 Epoch 00065: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1816e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 88ms/epoch - 5ms/step Epoch 66/500 Epoch 00066: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1760e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 88ms/epoch - 6ms/step Epoch 67/500 Epoch 00067: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1704e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 68/500 Epoch 00068: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1648e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step Epoch 69/500 Epoch 00069: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1591e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step Epoch 70/500 Epoch 00070: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1533e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 71/500 Epoch 00071: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1476e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step Epoch 72/500 Epoch 00072: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1418e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step Epoch 73/500 Epoch 00073: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1360e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step Epoch 74/500 Epoch 00074: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1301e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 93ms/epoch - 6ms/step Epoch 75/500 Epoch 00075: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1242e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step Epoch 76/500 Epoch 00076: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1183e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 77/500 Epoch 00077: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1123e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step Epoch 78/500 Epoch 00078: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1063e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step Epoch 79/500 Epoch 00079: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.1003e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 92ms/epoch - 6ms/step Epoch 80/500 Epoch 00080: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0942e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step Epoch 81/500 Epoch 00081: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0881e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step Epoch 82/500 Epoch 00082: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0820e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 89ms/epoch - 6ms/step Epoch 83/500 Epoch 00083: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0759e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 84/500 Epoch 00084: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0697e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step Epoch 85/500 Epoch 00085: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0635e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step Epoch 86/500 Epoch 00086: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0573e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step Epoch 87/500 Epoch 00087: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0510e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 88ms/epoch - 5ms/step Epoch 88/500 Epoch 00088: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0448e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 88ms/epoch - 6ms/step Epoch 89/500 Epoch 00089: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0384e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step Epoch 90/500 Epoch 00090: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0321e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 93ms/epoch - 6ms/step Epoch 91/500 Epoch 00091: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0258e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step Epoch 92/500 Epoch 00092: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0194e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step Epoch 93/500 Epoch 00093: val_loss did not improve from 0.00543 16/16 - 0s - loss: 8.0130e-04 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step Epoch 00093: early stopping
SMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 71.66891031373025
RMSE: 8.465749247038342
MAPE: 6.880610177712922
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 67.24711347278334
RMSE: 8.200433736869249
MAPE: 6.781803215137433
WMA
WMA([input_arrays], [timeperiod=30])
Weighted Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
49
Working on WMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-14480.432, Time=8.68 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-15747.905, Time=6.24 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15116.389, Time=6.89 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-13532.115, Time=7.94 sec
ARIMA(0,3,0)(0,0,0)[0] intercept : AIC=-13619.624, Time=5.24 sec
Best model: ARIMA(0,3,0)(0,0,0)[0]
Total fit time: 35.004 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(0, 3, 0) Log Likelihood 7895.952
Date: Sun, 12 Dec 2021 AIC -15747.905
Time: 16:48:05 BIC -15644.706
Sample: 0 HQIC -15708.272
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 3.384e-05 1.9e-05 1.778 0.075 -3.47e-06 7.12e-05
x2 3.379e-05 1.84e-05 1.832 0.067 -2.35e-06 6.99e-05
x3 3.388e-05 4.34e-05 0.781 0.435 -5.12e-05 0.000
x4 1.0000 4.12e-06 2.43e+05 0.000 1.000 1.000
x5 3.227e-05 3.52e-06 9.163 0.000 2.54e-05 3.92e-05
x6 5.559e-05 6.75e-05 0.823 0.410 -7.67e-05 0.000
x7 3.369e-05 2.38e-05 1.415 0.157 -1.3e-05 8.03e-05
x8 0.0023 2.6e-05 86.661 0.000 0.002 0.002
x9 -8.72e-06 7.51e-07 -11.610 0.000 -1.02e-05 -7.25e-06
x10 -0.0023 3.33e-05 -67.770 0.000 -0.002 -0.002
x11 0.0093 2.8e-05 333.459 0.000 0.009 0.009
x12 -0.0118 2.37e-05 -498.171 0.000 -0.012 -0.012
x13 3.382e-05 1.49e-05 2.273 0.023 4.66e-06 6.3e-05
x14 9.271e-05 6.21e-05 1.493 0.135 -2.9e-05 0.000
x15 3.096e-05 1.92e-05 1.614 0.106 -6.63e-06 6.86e-05
x16 5.52e-05 7.17e-05 0.770 0.441 -8.53e-05 0.000
x17 3.38e-05 3.2e-05 1.056 0.291 -2.89e-05 9.65e-05
x18 -6.715e-06 8.34e-05 -0.081 0.936 -0.000 0.000
x19 3.428e-05 2.07e-05 1.654 0.098 -6.34e-06 7.49e-05
x20 -8.089e-06 9.55e-05 -0.085 0.933 -0.000 0.000
x21 4.255e-05 0.000 0.094 0.925 -0.001 0.001
sigma2 2.581e-10 7.87e-11 3.280 0.001 1.04e-10 4.12e-10
===================================================================================
Ljung-Box (L1) (Q): 362.92 Jarque-Bera (JB): 5047564.68
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.01 Skew: -11.23
Prob(H) (two-sided): 0.00 Kurtosis: 390.28
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.75e+20. Standard errors may be unstable.
ARIMA order: (0, 3, 0)
/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead. super(Adam, self).__init__(name, **kwargs)
Epoch 1/500 Epoch 00001: val_loss improved from inf to 0.15122, saving model to LSTM6.h5 17/17 - 4s - loss: 0.1166 - accuracy: 0.0000e+00 - val_loss: 0.1512 - val_accuracy: 0.0000e+00 - lr: 0.0010 - 4s/epoch - 210ms/step Epoch 2/500 Epoch 00002: val_loss improved from 0.15122 to 0.08149, saving model to LSTM6.h5 17/17 - 0s - loss: 0.0463 - accuracy: 0.0000e+00 - val_loss: 0.0815 - val_accuracy: 0.0000e+00 - lr: 0.0010 - 124ms/epoch - 7ms/step Epoch 3/500 Epoch 00003: val_loss improved from 0.08149 to 0.00911, saving model to LSTM6.h5 17/17 - 0s - loss: 0.0137 - accuracy: 0.0000e+00 - val_loss: 0.0091 - val_accuracy: 0.0037 - lr: 0.0010 - 129ms/epoch - 8ms/step Epoch 4/500 Epoch 00004: val_loss improved from 0.00911 to 0.00792, saving model to LSTM6.h5 17/17 - 0s - loss: 0.0043 - accuracy: 0.0000e+00 - val_loss: 0.0079 - val_accuracy: 0.0037 - lr: 0.0010 - 113ms/epoch - 7ms/step Epoch 5/500 Epoch 00005: val_loss did not improve from 0.00792 17/17 - 0s - loss: 0.0047 - accuracy: 0.0000e+00 - val_loss: 0.0121 - val_accuracy: 0.0037 - lr: 0.0010 - 94ms/epoch - 6ms/step Epoch 6/500 Epoch 00006: val_loss did not improve from 0.00792 17/17 - 0s - loss: 0.0019 - accuracy: 0.0000e+00 - val_loss: 0.0160 - val_accuracy: 0.0037 - lr: 0.0010 - 99ms/epoch - 6ms/step Epoch 7/500 Epoch 00007: val_loss did not improve from 0.00792 17/17 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0155 - val_accuracy: 0.0037 - lr: 0.0010 - 90ms/epoch - 5ms/step Epoch 8/500 Epoch 00008: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.8841e-04 - accuracy: 0.0000e+00 - val_loss: 0.0143 - val_accuracy: 0.0037 - lr: 0.0010 - 94ms/epoch - 6ms/step Epoch 9/500 Epoch 00009: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513. Epoch 00009: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.5937e-04 - accuracy: 0.0000e+00 - val_loss: 0.0142 - val_accuracy: 0.0037 - lr: 0.0010 - 94ms/epoch - 6ms/step Epoch 10/500 Epoch 00010: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.3381e-04 - accuracy: 0.0000e+00 - val_loss: 0.0137 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 91ms/epoch - 5ms/step Epoch 11/500 Epoch 00011: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.1282e-04 - accuracy: 0.0000e+00 - val_loss: 0.0138 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 99ms/epoch - 6ms/step Epoch 12/500 Epoch 00012: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.1136e-04 - accuracy: 0.0000e+00 - val_loss: 0.0141 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 106ms/epoch - 6ms/step Epoch 13/500 Epoch 00013: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0814e-04 - accuracy: 0.0000e+00 - val_loss: 0.0142 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 104ms/epoch - 6ms/step Epoch 14/500 Epoch 00014: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05. Epoch 00014: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0537e-04 - accuracy: 0.0000e+00 - val_loss: 0.0143 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 99ms/epoch - 6ms/step Epoch 15/500 Epoch 00015: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0338e-04 - accuracy: 0.0000e+00 - val_loss: 0.0143 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 16/500 Epoch 00016: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0311e-04 - accuracy: 0.0000e+00 - val_loss: 0.0144 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step Epoch 17/500 Epoch 00017: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0283e-04 - accuracy: 0.0000e+00 - val_loss: 0.0144 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step Epoch 18/500 Epoch 00018: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0256e-04 - accuracy: 0.0000e+00 - val_loss: 0.0144 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 19/500 Epoch 00019: ReduceLROnPlateau reducing learning rate to 1e-05. Epoch 00019: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0228e-04 - accuracy: 0.0000e+00 - val_loss: 0.0144 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 90ms/epoch - 5ms/step Epoch 20/500 Epoch 00020: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0200e-04 - accuracy: 0.0000e+00 - val_loss: 0.0144 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 21/500 Epoch 00021: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0171e-04 - accuracy: 0.0000e+00 - val_loss: 0.0144 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 105ms/epoch - 6ms/step Epoch 22/500 Epoch 00022: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0142e-04 - accuracy: 0.0000e+00 - val_loss: 0.0144 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 98ms/epoch - 6ms/step Epoch 23/500 Epoch 00023: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0111e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 102ms/epoch - 6ms/step Epoch 24/500 Epoch 00024: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0081e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 102ms/epoch - 6ms/step Epoch 25/500 Epoch 00025: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0049e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 101ms/epoch - 6ms/step Epoch 26/500 Epoch 00026: val_loss did not improve from 0.00792 17/17 - 0s - loss: 9.0017e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step Epoch 27/500 Epoch 00027: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9984e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step Epoch 28/500 Epoch 00028: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9950e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step Epoch 29/500 Epoch 00029: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9916e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 98ms/epoch - 6ms/step Epoch 30/500 Epoch 00030: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9881e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step Epoch 31/500 Epoch 00031: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9845e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 104ms/epoch - 6ms/step Epoch 32/500 Epoch 00032: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9809e-04 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 103ms/epoch - 6ms/step Epoch 33/500 Epoch 00033: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9772e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step Epoch 34/500 Epoch 00034: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9735e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 102ms/epoch - 6ms/step Epoch 35/500 Epoch 00035: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9697e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 90ms/epoch - 5ms/step Epoch 36/500 Epoch 00036: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9658e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 90ms/epoch - 5ms/step Epoch 37/500 Epoch 00037: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9619e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 102ms/epoch - 6ms/step Epoch 38/500 Epoch 00038: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9579e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 100ms/epoch - 6ms/step Epoch 39/500 Epoch 00039: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9539e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step Epoch 40/500 Epoch 00040: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9498e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 103ms/epoch - 6ms/step Epoch 41/500 Epoch 00041: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9457e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 102ms/epoch - 6ms/step Epoch 42/500 Epoch 00042: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9415e-04 - accuracy: 0.0000e+00 - val_loss: 0.0146 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step Epoch 43/500 Epoch 00043: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9373e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step Epoch 44/500 Epoch 00044: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9330e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step Epoch 45/500 Epoch 00045: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9286e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step Epoch 46/500 Epoch 00046: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9242e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step Epoch 47/500 Epoch 00047: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9197e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step Epoch 48/500 Epoch 00048: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9152e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step Epoch 49/500 Epoch 00049: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9107e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 50/500 Epoch 00050: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9061e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 106ms/epoch - 6ms/step Epoch 51/500 Epoch 00051: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.9014e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 100ms/epoch - 6ms/step Epoch 52/500 Epoch 00052: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.8967e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step Epoch 53/500 Epoch 00053: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.8919e-04 - accuracy: 0.0000e+00 - val_loss: 0.0147 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 102ms/epoch - 6ms/step Epoch 54/500 Epoch 00054: val_loss did not improve from 0.00792 17/17 - 0s - loss: 8.8871e-04 - accuracy: 0.0000e+00 - val_loss: 0.0148 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 102ms/epoch - 6ms/step Epoch 00054: early stopping
SMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 71.66891031373025
RMSE: 8.465749247038342
MAPE: 6.880610177712922
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 67.24711347278334
RMSE: 8.200433736869249
MAPE: 6.781803215137433
WMA
Prediction vs Close: 55.6% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 58.550384022767716
RMSE: 7.6518222681115455
MAPE: 6.1413991074844
DEMA
DEMA([input_arrays], [timeperiod=30])
Double Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
89
Working on DEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.774, Time=2.34 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=4.01 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15590.302, Time=6.94 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.65 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15269.503, Time=7.14 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-16414.961, Time=8.11 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16878.396, Time=9.36 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17030.019, Time=3.97 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.613, Time=2.81 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.441, Time=5.86 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=14.61 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17003.985, Time=3.09 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-16998.665, Time=3.65 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 77.572 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.721
Date: Sun, 12 Dec 2021 AIC -17087.441
Time: 16:49:55 BIC -16965.479
Sample: 0 HQIC -17040.603
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.799e-10 1.43e-20 -1.96e+10 0.000 -2.8e-10 -2.8e-10
x2 -2.817e-10 1.43e-20 -1.97e+10 0.000 -2.82e-10 -2.82e-10
x3 -2.805e-10 1.43e-20 -1.96e+10 0.000 -2.8e-10 -2.8e-10
x4 1.0000 1.43e-20 7e+19 0.000 1.000 1.000
x5 -2.597e-10 1.37e-20 -1.89e+10 0.000 -2.6e-10 -2.6e-10
x6 -1.388e-09 3.12e-20 -4.45e+10 0.000 -1.39e-09 -1.39e-09
x7 -2.789e-10 1.42e-20 -1.96e+10 0.000 -2.79e-10 -2.79e-10
x8 -2.76e-10 1.42e-20 -1.95e+10 0.000 -2.76e-10 -2.76e-10
x9 -2.216e-12 3.53e-22 -6.28e+09 0.000 -2.22e-12 -2.22e-12
x10 -1.345e-10 9.82e-21 -1.37e+10 0.000 -1.34e-10 -1.34e-10
x11 -2.898e-10 1.45e-20 -2e+10 0.000 -2.9e-10 -2.9e-10
x12 -2.602e-10 1.38e-20 -1.89e+10 0.000 -2.6e-10 -2.6e-10
x13 -2.807e-10 1.43e-20 -1.96e+10 0.000 -2.81e-10 -2.81e-10
x14 -1.87e-09 3.69e-20 -5.07e+10 0.000 -1.87e-09 -1.87e-09
x15 -2.726e-10 1.43e-20 -1.9e+10 0.000 -2.73e-10 -2.73e-10
x16 -7.915e-11 7.68e-21 -1.03e+10 0.000 -7.92e-11 -7.92e-11
x17 -2.606e-10 1.33e-20 -1.96e+10 0.000 -2.61e-10 -2.61e-10
x18 -6.408e-10 2.16e-20 -2.97e+10 0.000 -6.41e-10 -6.41e-10
x19 -2.881e-10 1.46e-20 -1.98e+10 0.000 -2.88e-10 -2.88e-10
x20 -4.337e-10 1.78e-20 -2.44e+10 0.000 -4.34e-10 -4.34e-10
x21 -4.549e-10 1.79e-20 -2.55e+10 0.000 -4.55e-10 -4.55e-10
ar.L1 -0.4923 1.46e-22 -3.38e+21 0.000 -0.492 -0.492
ar.L2 -0.1923 8.47e-23 -2.27e+21 0.000 -0.192 -0.192
ar.L3 -0.0461 4.02e-23 -1.15e+21 0.000 -0.046 -0.046
ma.L1 -0.7078 3.31e-22 -2.14e+21 0.000 -0.708 -0.708
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 55.07 Jarque-Bera (JB): 4171695.82
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.26
Prob(H) (two-sided): 0.00 Kurtosis: 355.51
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 2.62e+41. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead. super(Adam, self).__init__(name, **kwargs)
Epoch 1/500 Epoch 00001: val_loss improved from inf to 0.06618, saving model to LSTM6.h5 10/10 - 4s - loss: 0.2504 - accuracy: 0.0000e+00 - val_loss: 0.0662 - val_accuracy: 0.0000e+00 - lr: 0.0010 - 4s/epoch - 380ms/step Epoch 2/500 Epoch 00002: val_loss improved from 0.06618 to 0.03100, saving model to LSTM6.h5 10/10 - 0s - loss: 0.0628 - accuracy: 0.0000e+00 - val_loss: 0.0310 - val_accuracy: 0.0037 - lr: 0.0010 - 91ms/epoch - 9ms/step Epoch 3/500 Epoch 00003: val_loss did not improve from 0.03100 10/10 - 0s - loss: 0.0124 - accuracy: 0.0000e+00 - val_loss: 0.0482 - val_accuracy: 0.0037 - lr: 0.0010 - 63ms/epoch - 6ms/step Epoch 4/500 Epoch 00004: val_loss did not improve from 0.03100 10/10 - 0s - loss: 0.0035 - accuracy: 0.0000e+00 - val_loss: 0.0460 - val_accuracy: 0.0037 - lr: 0.0010 - 66ms/epoch - 7ms/step Epoch 5/500 Epoch 00005: val_loss did not improve from 0.03100 10/10 - 0s - loss: 0.0018 - accuracy: 0.0000e+00 - val_loss: 0.0350 - val_accuracy: 0.0037 - lr: 0.0010 - 62ms/epoch - 6ms/step Epoch 6/500 Epoch 00006: val_loss improved from 0.03100 to 0.02818, saving model to LSTM6.h5 10/10 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0282 - val_accuracy: 0.0037 - lr: 0.0010 - 86ms/epoch - 9ms/step Epoch 7/500 Epoch 00007: val_loss improved from 0.02818 to 0.02201, saving model to LSTM6.h5 10/10 - 0s - loss: 9.9207e-04 - accuracy: 0.0000e+00 - val_loss: 0.0220 - val_accuracy: 0.0037 - lr: 0.0010 - 77ms/epoch - 8ms/step Epoch 8/500 Epoch 00008: val_loss did not improve from 0.02201 10/10 - 0s - loss: 9.8483e-04 - accuracy: 0.0000e+00 - val_loss: 0.0234 - val_accuracy: 0.0037 - lr: 0.0010 - 59ms/epoch - 6ms/step Epoch 9/500 Epoch 00009: val_loss did not improve from 0.02201 10/10 - 0s - loss: 9.3177e-04 - accuracy: 0.0000e+00 - val_loss: 0.0234 - val_accuracy: 0.0037 - lr: 0.0010 - 63ms/epoch - 6ms/step Epoch 10/500 Epoch 00010: val_loss did not improve from 0.02201 10/10 - 0s - loss: 9.8743e-04 - accuracy: 0.0000e+00 - val_loss: 0.0261 - val_accuracy: 0.0037 - lr: 0.0010 - 64ms/epoch - 6ms/step Epoch 11/500 Epoch 00011: val_loss did not improve from 0.02201 10/10 - 0s - loss: 9.0889e-04 - accuracy: 0.0000e+00 - val_loss: 0.0251 - val_accuracy: 0.0037 - lr: 0.0010 - 67ms/epoch - 7ms/step Epoch 12/500 Epoch 00012: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513. Epoch 00012: val_loss did not improve from 0.02201 10/10 - 0s - loss: 9.3831e-04 - accuracy: 0.0000e+00 - val_loss: 0.0258 - val_accuracy: 0.0037 - lr: 0.0010 - 69ms/epoch - 7ms/step Epoch 13/500 Epoch 00013: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.8311e-04 - accuracy: 0.0000e+00 - val_loss: 0.0258 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 68ms/epoch - 7ms/step Epoch 14/500 Epoch 00014: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.8204e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 60ms/epoch - 6ms/step Epoch 15/500 Epoch 00015: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.8051e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 68ms/epoch - 7ms/step Epoch 16/500 Epoch 00016: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7889e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 63ms/epoch - 6ms/step Epoch 17/500 Epoch 00017: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05. Epoch 00017: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7749e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 65ms/epoch - 6ms/step Epoch 18/500 Epoch 00018: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7558e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 62ms/epoch - 6ms/step Epoch 19/500 Epoch 00019: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7549e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step Epoch 20/500 Epoch 00020: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7538e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 64ms/epoch - 6ms/step Epoch 21/500 Epoch 00021: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7526e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 70ms/epoch - 7ms/step Epoch 22/500 Epoch 00022: ReduceLROnPlateau reducing learning rate to 1e-05. Epoch 00022: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7513e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 65ms/epoch - 6ms/step Epoch 23/500 Epoch 00023: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7501e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step Epoch 24/500 Epoch 00024: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7488e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 70ms/epoch - 7ms/step Epoch 25/500 Epoch 00025: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7474e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step Epoch 26/500 Epoch 00026: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7460e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step Epoch 27/500 Epoch 00027: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7446e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 74ms/epoch - 7ms/step Epoch 28/500 Epoch 00028: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7432e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 67ms/epoch - 7ms/step Epoch 29/500 Epoch 00029: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7417e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step Epoch 30/500 Epoch 00030: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7402e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 70ms/epoch - 7ms/step Epoch 31/500 Epoch 00031: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7387e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step Epoch 32/500 Epoch 00032: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7371e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 64ms/epoch - 6ms/step Epoch 33/500 Epoch 00033: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7356e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 71ms/epoch - 7ms/step Epoch 34/500 Epoch 00034: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7340e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step Epoch 35/500 Epoch 00035: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7323e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step Epoch 36/500 Epoch 00036: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7307e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 65ms/epoch - 7ms/step Epoch 37/500 Epoch 00037: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7290e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 65ms/epoch - 7ms/step Epoch 38/500 Epoch 00038: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7273e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step Epoch 39/500 Epoch 00039: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7256e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step Epoch 40/500 Epoch 00040: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7238e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step Epoch 41/500 Epoch 00041: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7221e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 64ms/epoch - 6ms/step Epoch 42/500 Epoch 00042: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7203e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step Epoch 43/500 Epoch 00043: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7185e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 62ms/epoch - 6ms/step Epoch 44/500 Epoch 00044: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7167e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 62ms/epoch - 6ms/step Epoch 45/500 Epoch 00045: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7149e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 65ms/epoch - 6ms/step Epoch 46/500 Epoch 00046: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7130e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 67ms/epoch - 7ms/step Epoch 47/500 Epoch 00047: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7111e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step Epoch 48/500 Epoch 00048: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7092e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 72ms/epoch - 7ms/step Epoch 49/500 Epoch 00049: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7073e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step Epoch 50/500 Epoch 00050: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7054e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step Epoch 51/500 Epoch 00051: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7035e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step Epoch 52/500 Epoch 00052: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.7015e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 73ms/epoch - 7ms/step Epoch 53/500 Epoch 00053: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.6995e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step Epoch 54/500 Epoch 00054: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.6975e-04 - accuracy: 0.0000e+00 - val_loss: 0.0259 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 69ms/epoch - 7ms/step Epoch 55/500 Epoch 00055: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.6955e-04 - accuracy: 0.0000e+00 - val_loss: 0.0260 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 64ms/epoch - 6ms/step Epoch 56/500 Epoch 00056: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.6935e-04 - accuracy: 0.0000e+00 - val_loss: 0.0260 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step Epoch 57/500 Epoch 00057: val_loss did not improve from 0.02201 10/10 - 0s - loss: 8.6915e-04 - accuracy: 0.0000e+00 - val_loss: 0.0260 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step Epoch 00057: early stopping
SMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 71.66891031373025
RMSE: 8.465749247038342
MAPE: 6.880610177712922
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 67.24711347278334
RMSE: 8.200433736869249
MAPE: 6.781803215137433
WMA
Prediction vs Close: 55.6% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 58.550384022767716
RMSE: 7.6518222681115455
MAPE: 6.1413991074844
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 127.72881036471918
RMSE: 11.301717142307146
MAPE: 10.306940424406019
KAMA
KAMA([input_arrays], [timeperiod=30])
Kaufman Adaptive Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
18
Working on KAMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.902, Time=2.09 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=3.81 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16796.316, Time=7.94 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.24 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-17004.193, Time=2.14 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15176.063, Time=9.59 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16873.638, Time=9.98 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17008.756, Time=2.44 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.764, Time=3.39 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-15723.849, Time=12.69 sec
ARIMA(2,3,0)(0,0,0)[0] intercept : AIC=-17006.756, Time=2.45 sec
Best model: ARIMA(2,3,0)(0,0,0)[0]
Total fit time: 61.772 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(2, 3, 0) Log Likelihood 8528.378
Date: Sun, 12 Dec 2021 AIC -17008.756
Time: 16:58:40 BIC -16896.176
Sample: 0 HQIC -16965.520
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.24e-15 7.41e-26 -3.02e+10 0.000 -2.24e-15 -2.24e-15
x2 8.461e-16 6.6e-26 1.28e+10 0.000 8.46e-16 8.46e-16
x3 4.901e-16 6.89e-26 7.11e+09 0.000 4.9e-16 4.9e-16
x4 1.0000 6.96e-26 1.44e+25 0.000 1.000 1.000
x5 5.931e-15 6.61e-26 8.97e+10 0.000 5.93e-15 5.93e-15
x6 -1.05e-15 1.5e-25 -7e+09 0.000 -1.05e-15 -1.05e-15
x7 1.439e-15 6.87e-26 2.1e+10 0.000 1.44e-15 1.44e-15
x8 -1.25e-15 6.8e-26 -1.84e+10 0.000 -1.25e-15 -1.25e-15
x9 -9.356e-17 8.97e-27 -1.04e+10 0.000 -9.36e-17 -9.36e-17
x10 -1.145e-16 2.88e-26 -3.98e+09 0.000 -1.15e-16 -1.15e-16
x11 -2.036e-16 6.8e-26 -3e+09 0.000 -2.04e-16 -2.04e-16
x12 5.951e-16 6.76e-26 8.81e+09 0.000 5.95e-16 5.95e-16
x13 -6.117e-17 6.94e-26 -8.81e+08 0.000 -6.12e-17 -6.12e-17
x14 1.167e-15 1.99e-25 5.85e+09 0.000 1.17e-15 1.17e-15
x15 -4.274e-14 6.99e-26 -6.11e+11 0.000 -4.27e-14 -4.27e-14
x16 2.262e-14 8.56e-26 2.64e+11 0.000 2.26e-14 2.26e-14
x17 3.384e-14 6.46e-26 5.24e+11 0.000 3.38e-14 3.38e-14
x18 9.894e-16 5.8e-26 1.71e+10 0.000 9.89e-16 9.89e-16
x19 4.115e-14 7.75e-26 5.31e+11 0.000 4.12e-14 4.12e-14
x20 -2.176e-15 9.49e-26 -2.29e+10 0.000 -2.18e-15 -2.18e-15
x21 -7.755e-17 4.63e-26 -1.67e+09 0.000 -7.75e-17 -7.75e-17
ar.L1 -0.9988 9.76e-22 -1.02e+21 0.000 -0.999 -0.999
ar.L2 -0.4972 4.07e-23 -1.22e+22 0.000 -0.497 -0.497
sigma2 1e-10 6.99e-11 1.432 0.152 -3.69e-11 2.37e-10
===================================================================================
Ljung-Box (L1) (Q): 31.54 Jarque-Bera (JB): 2432532.03
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: -0.15
Prob(H) (two-sided): 0.00 Kurtosis: 272.30
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 7.19e+48. Standard errors may be unstable.
ARIMA order: (2, 3, 0)
/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead. super(Adam, self).__init__(name, **kwargs)
Epoch 1/500 Epoch 00001: val_loss improved from inf to 0.04322, saving model to LSTM6.h5 45/45 - 4s - loss: 0.1157 - accuracy: 0.0000e+00 - val_loss: 0.0432 - val_accuracy: 0.0037 - lr: 0.0010 - 4s/epoch - 88ms/step Epoch 2/500 Epoch 00002: val_loss improved from 0.04322 to 0.02118, saving model to LSTM6.h5 45/45 - 0s - loss: 0.0138 - accuracy: 0.0000e+00 - val_loss: 0.0212 - val_accuracy: 0.0037 - lr: 0.0010 - 223ms/epoch - 5ms/step Epoch 3/500 Epoch 00003: val_loss did not improve from 0.02118 45/45 - 0s - loss: 0.0091 - accuracy: 0.0000e+00 - val_loss: 0.0243 - val_accuracy: 0.0037 - lr: 0.0010 - 220ms/epoch - 5ms/step Epoch 4/500 Epoch 00004: val_loss improved from 0.02118 to 0.00757, saving model to LSTM6.h5 45/45 - 0s - loss: 0.0124 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 0.0010 - 249ms/epoch - 6ms/step Epoch 5/500 Epoch 00005: val_loss did not improve from 0.00757 45/45 - 0s - loss: 0.0094 - accuracy: 0.0000e+00 - val_loss: 0.0082 - val_accuracy: 0.0037 - lr: 0.0010 - 214ms/epoch - 5ms/step Epoch 6/500 Epoch 00006: val_loss improved from 0.00757 to 0.00671, saving model to LSTM6.h5 45/45 - 0s - loss: 0.0145 - accuracy: 0.0000e+00 - val_loss: 0.0067 - val_accuracy: 0.0037 - lr: 0.0010 - 232ms/epoch - 5ms/step Epoch 7/500 Epoch 00007: val_loss improved from 0.00671 to 0.00595, saving model to LSTM6.h5 45/45 - 0s - loss: 0.0087 - accuracy: 0.0000e+00 - val_loss: 0.0060 - val_accuracy: 0.0037 - lr: 0.0010 - 228ms/epoch - 5ms/step Epoch 8/500 Epoch 00008: val_loss improved from 0.00595 to 0.00564, saving model to LSTM6.h5 45/45 - 0s - loss: 0.0035 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 0.0010 - 231ms/epoch - 5ms/step Epoch 9/500 Epoch 00009: val_loss did not improve from 0.00564 45/45 - 0s - loss: 0.0015 - accuracy: 0.0000e+00 - val_loss: 0.0064 - val_accuracy: 0.0037 - lr: 0.0010 - 218ms/epoch - 5ms/step Epoch 10/500 Epoch 00010: val_loss did not improve from 0.00564 45/45 - 0s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0071 - val_accuracy: 0.0037 - lr: 0.0010 - 215ms/epoch - 5ms/step Epoch 11/500 Epoch 00011: val_loss did not improve from 0.00564 45/45 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0089 - val_accuracy: 0.0037 - lr: 0.0010 - 219ms/epoch - 5ms/step Epoch 12/500 Epoch 00012: val_loss did not improve from 0.00564 45/45 - 0s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0098 - val_accuracy: 0.0037 - lr: 0.0010 - 230ms/epoch - 5ms/step Epoch 13/500 Epoch 00013: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513. Epoch 00013: val_loss did not improve from 0.00564 45/45 - 0s - loss: 0.0025 - accuracy: 0.0000e+00 - val_loss: 0.0134 - val_accuracy: 0.0037 - lr: 0.0010 - 217ms/epoch - 5ms/step Epoch 14/500 Epoch 00014: val_loss did not improve from 0.00564 45/45 - 0s - loss: 0.0050 - accuracy: 0.0000e+00 - val_loss: 0.0070 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 211ms/epoch - 5ms/step Epoch 15/500 Epoch 00015: val_loss did not improve from 0.00564 45/45 - 0s - loss: 0.0016 - accuracy: 0.0000e+00 - val_loss: 0.0082 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 203ms/epoch - 5ms/step Epoch 16/500 Epoch 00016: val_loss did not improve from 0.00564 45/45 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0090 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 212ms/epoch - 5ms/step Epoch 17/500 Epoch 00017: val_loss did not improve from 0.00564 45/45 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0098 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 221ms/epoch - 5ms/step Epoch 18/500 Epoch 00018: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05. Epoch 00018: val_loss did not improve from 0.00564 45/45 - 0s - loss: 9.2074e-04 - accuracy: 0.0000e+00 - val_loss: 0.0105 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 206ms/epoch - 5ms/step Epoch 19/500 Epoch 00019: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.5252e-04 - accuracy: 0.0000e+00 - val_loss: 0.0106 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step Epoch 20/500 Epoch 00020: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.4270e-04 - accuracy: 0.0000e+00 - val_loss: 0.0108 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 208ms/epoch - 5ms/step Epoch 21/500 Epoch 00021: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.3744e-04 - accuracy: 0.0000e+00 - val_loss: 0.0109 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step Epoch 22/500 Epoch 00022: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.3412e-04 - accuracy: 0.0000e+00 - val_loss: 0.0109 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step Epoch 23/500 Epoch 00023: ReduceLROnPlateau reducing learning rate to 1e-05. Epoch 00023: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.3146e-04 - accuracy: 0.0000e+00 - val_loss: 0.0110 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 204ms/epoch - 5ms/step Epoch 24/500 Epoch 00024: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.2904e-04 - accuracy: 0.0000e+00 - val_loss: 0.0110 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step Epoch 25/500 Epoch 00025: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.2675e-04 - accuracy: 0.0000e+00 - val_loss: 0.0111 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step Epoch 26/500 Epoch 00026: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.2453e-04 - accuracy: 0.0000e+00 - val_loss: 0.0111 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 234ms/epoch - 5ms/step Epoch 27/500 Epoch 00027: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.2238e-04 - accuracy: 0.0000e+00 - val_loss: 0.0112 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step Epoch 28/500 Epoch 00028: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.2029e-04 - accuracy: 0.0000e+00 - val_loss: 0.0112 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 203ms/epoch - 5ms/step Epoch 29/500 Epoch 00029: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.1825e-04 - accuracy: 0.0000e+00 - val_loss: 0.0113 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step Epoch 30/500 Epoch 00030: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.1626e-04 - accuracy: 0.0000e+00 - val_loss: 0.0113 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step Epoch 31/500 Epoch 00031: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.1433e-04 - accuracy: 0.0000e+00 - val_loss: 0.0114 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 242ms/epoch - 5ms/step Epoch 32/500 Epoch 00032: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.1244e-04 - accuracy: 0.0000e+00 - val_loss: 0.0114 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step Epoch 33/500 Epoch 00033: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.1061e-04 - accuracy: 0.0000e+00 - val_loss: 0.0115 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step Epoch 34/500 Epoch 00034: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.0883e-04 - accuracy: 0.0000e+00 - val_loss: 0.0115 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 215ms/epoch - 5ms/step Epoch 35/500 Epoch 00035: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.0709e-04 - accuracy: 0.0000e+00 - val_loss: 0.0115 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step Epoch 36/500 Epoch 00036: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.0540e-04 - accuracy: 0.0000e+00 - val_loss: 0.0116 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step Epoch 37/500 Epoch 00037: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.0376e-04 - accuracy: 0.0000e+00 - val_loss: 0.0116 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step Epoch 38/500 Epoch 00038: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.0216e-04 - accuracy: 0.0000e+00 - val_loss: 0.0117 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step Epoch 39/500 Epoch 00039: val_loss did not improve from 0.00564 45/45 - 0s - loss: 8.0060e-04 - accuracy: 0.0000e+00 - val_loss: 0.0117 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step Epoch 40/500 Epoch 00040: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.9908e-04 - accuracy: 0.0000e+00 - val_loss: 0.0117 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 237ms/epoch - 5ms/step Epoch 41/500 Epoch 00041: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.9760e-04 - accuracy: 0.0000e+00 - val_loss: 0.0118 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 230ms/epoch - 5ms/step Epoch 42/500 Epoch 00042: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.9615e-04 - accuracy: 0.0000e+00 - val_loss: 0.0118 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step Epoch 43/500 Epoch 00043: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.9473e-04 - accuracy: 0.0000e+00 - val_loss: 0.0119 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step Epoch 44/500 Epoch 00044: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.9334e-04 - accuracy: 0.0000e+00 - val_loss: 0.0119 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step Epoch 45/500 Epoch 00045: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.9198e-04 - accuracy: 0.0000e+00 - val_loss: 0.0119 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step Epoch 46/500 Epoch 00046: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.9065e-04 - accuracy: 0.0000e+00 - val_loss: 0.0119 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step Epoch 47/500 Epoch 00047: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.8934e-04 - accuracy: 0.0000e+00 - val_loss: 0.0120 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step Epoch 48/500 Epoch 00048: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.8805e-04 - accuracy: 0.0000e+00 - val_loss: 0.0120 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step Epoch 49/500 Epoch 00049: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.8678e-04 - accuracy: 0.0000e+00 - val_loss: 0.0120 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 235ms/epoch - 5ms/step Epoch 50/500 Epoch 00050: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.8553e-04 - accuracy: 0.0000e+00 - val_loss: 0.0120 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 205ms/epoch - 5ms/step Epoch 51/500 Epoch 00051: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.8430e-04 - accuracy: 0.0000e+00 - val_loss: 0.0121 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step Epoch 52/500 Epoch 00052: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.8308e-04 - accuracy: 0.0000e+00 - val_loss: 0.0121 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step Epoch 53/500 Epoch 00053: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.8187e-04 - accuracy: 0.0000e+00 - val_loss: 0.0121 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step Epoch 54/500 Epoch 00054: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.8068e-04 - accuracy: 0.0000e+00 - val_loss: 0.0121 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step Epoch 55/500 Epoch 00055: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.7949e-04 - accuracy: 0.0000e+00 - val_loss: 0.0121 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 202ms/epoch - 4ms/step Epoch 56/500 Epoch 00056: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.7832e-04 - accuracy: 0.0000e+00 - val_loss: 0.0121 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step Epoch 57/500 Epoch 00057: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.7716e-04 - accuracy: 0.0000e+00 - val_loss: 0.0122 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step Epoch 58/500 Epoch 00058: val_loss did not improve from 0.00564 45/45 - 0s - loss: 7.7600e-04 - accuracy: 0.0000e+00 - val_loss: 0.0122 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step Epoch 00058: early stopping
SMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 71.66891031373025
RMSE: 8.465749247038342
MAPE: 6.880610177712922
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 67.24711347278334
RMSE: 8.200433736869249
MAPE: 6.781803215137433
WMA
Prediction vs Close: 55.6% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 58.550384022767716
RMSE: 7.6518222681115455
MAPE: 6.1413991074844
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 127.72881036471918
RMSE: 11.301717142307146
MAPE: 10.306940424406019
KAMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 48.51% Accuracy
MSE: 43.855468679122
RMSE: 6.622346161227303
MAPE: 5.4751276749367985
MIDPOINT
MIDPOINT([input_arrays], [timeperiod=14])
MidPoint over period (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 14
Outputs:
real
14
Working on MIDPOINT predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.753, Time=2.12 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.592, Time=3.89 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16288.639, Time=10.65 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.592, Time=5.03 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15275.254, Time=7.35 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15486.751, Time=12.13 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=48.000, Time=0.45 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17008.491, Time=2.27 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.554, Time=2.87 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.445, Time=5.71 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=-15686.421, Time=9.73 sec
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17030.168, Time=14.06 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-15138.715, Time=13.34 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 89.606 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.722
Date: Sun, 12 Dec 2021 AIC -17087.445
Time: 17:01:17 BIC -16965.483
Sample: 0 HQIC -17040.607
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.14e-10 1.09e-20 -1.96e+10 0.000 -2.14e-10 -2.14e-10
x2 -2.126e-10 1.13e-20 -1.88e+10 0.000 -2.13e-10 -2.13e-10
x3 -2.175e-10 1.06e-20 -2.04e+10 0.000 -2.17e-10 -2.17e-10
x4 1.0000 1.1e-20 9.11e+19 0.000 1.000 1.000
x5 -1.941e-10 1.05e-20 -1.86e+10 0.000 -1.94e-10 -1.94e-10
x6 -4.131e-09 7.64e-20 -5.4e+10 0.000 -4.13e-09 -4.13e-09
x7 -1.965e-10 1.05e-20 -1.86e+10 0.000 -1.96e-10 -1.96e-10
x8 -1.961e-10 1.07e-20 -1.84e+10 0.000 -1.96e-10 -1.96e-10
x9 -1.005e-10 9.12e-22 -1.1e+11 0.000 -1e-10 -1e-10
x10 -1.739e-10 3.37e-21 -5.16e+10 0.000 -1.74e-10 -1.74e-10
x11 -1.941e-10 1.07e-20 -1.82e+10 0.000 -1.94e-10 -1.94e-10
x12 -2.005e-10 1.06e-20 -1.89e+10 0.000 -2e-10 -2e-10
x13 -2.056e-10 1.07e-20 -1.91e+10 0.000 -2.06e-10 -2.06e-10
x14 -1.687e-09 3.15e-20 -5.36e+10 0.000 -1.69e-09 -1.69e-09
x15 -2.365e-10 1.17e-20 -2.01e+10 0.000 -2.36e-10 -2.36e-10
x16 -1.523e-10 9.42e-21 -1.62e+10 0.000 -1.52e-10 -1.52e-10
x17 -1.491e-10 9.33e-21 -1.6e+10 0.000 -1.49e-10 -1.49e-10
x18 -6.404e-10 1.93e-20 -3.32e+10 0.000 -6.4e-10 -6.4e-10
x19 -2.596e-10 1.23e-20 -2.11e+10 0.000 -2.6e-10 -2.6e-10
x20 -6.246e-10 1.91e-20 -3.28e+10 0.000 -6.25e-10 -6.25e-10
x21 -1.953e-09 2.16e-20 -9.04e+10 0.000 -1.95e-09 -1.95e-09
ar.L1 -0.4914 1.46e-22 -3.38e+21 0.000 -0.491 -0.491
ar.L2 -0.1934 8.48e-23 -2.28e+21 0.000 -0.193 -0.193
ar.L3 -0.0491 4.2e-23 -1.17e+21 0.000 -0.049 -0.049
ma.L1 -0.7092 3.33e-22 -2.13e+21 0.000 -0.709 -0.709
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 32.51 Jarque-Bera (JB): 49038.38
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.35 Skew: 1.06
Prob(H) (two-sided): 0.00 Kurtosis: 41.18
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.71e+40. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead. super(Adam, self).__init__(name, **kwargs)
Epoch 1/500 Epoch 00001: val_loss improved from inf to 0.00811, saving model to LSTM6.h5 58/58 - 4s - loss: 0.1472 - accuracy: 0.0000e+00 - val_loss: 0.0081 - val_accuracy: 0.0037 - lr: 0.0010 - 4s/epoch - 63ms/step Epoch 2/500 Epoch 00002: val_loss did not improve from 0.00811 58/58 - 0s - loss: 0.0220 - accuracy: 0.0000e+00 - val_loss: 0.0081 - val_accuracy: 0.0037 - lr: 0.0010 - 293ms/epoch - 5ms/step Epoch 3/500 Epoch 00003: val_loss did not improve from 0.00811 58/58 - 0s - loss: 0.0168 - accuracy: 0.0000e+00 - val_loss: 0.0433 - val_accuracy: 0.0037 - lr: 0.0010 - 274ms/epoch - 5ms/step Epoch 4/500 Epoch 00004: val_loss improved from 0.00811 to 0.00638, saving model to LSTM6.h5 58/58 - 0s - loss: 0.0052 - accuracy: 0.0000e+00 - val_loss: 0.0064 - val_accuracy: 0.0037 - lr: 0.0010 - 299ms/epoch - 5ms/step Epoch 5/500 Epoch 00005: val_loss did not improve from 0.00638 58/58 - 0s - loss: 0.0028 - accuracy: 0.0000e+00 - val_loss: 0.0088 - val_accuracy: 0.0037 - lr: 0.0010 - 273ms/epoch - 5ms/step Epoch 6/500 Epoch 00006: val_loss did not improve from 0.00638 58/58 - 0s - loss: 0.0062 - accuracy: 0.0000e+00 - val_loss: 0.0069 - val_accuracy: 0.0037 - lr: 0.0010 - 283ms/epoch - 5ms/step Epoch 7/500 Epoch 00007: val_loss did not improve from 0.00638 58/58 - 0s - loss: 0.0116 - accuracy: 0.0000e+00 - val_loss: 0.0070 - val_accuracy: 0.0037 - lr: 0.0010 - 271ms/epoch - 5ms/step Epoch 8/500 Epoch 00008: val_loss improved from 0.00638 to 0.00531, saving model to LSTM6.h5 58/58 - 0s - loss: 0.0209 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 0.0010 - 281ms/epoch - 5ms/step Epoch 9/500 Epoch 00009: val_loss improved from 0.00531 to 0.00390, saving model to LSTM6.h5 58/58 - 0s - loss: 0.0246 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 0.0010 - 299ms/epoch - 5ms/step Epoch 10/500 Epoch 00010: val_loss did not improve from 0.00390 58/58 - 0s - loss: 0.0153 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 0.0010 - 275ms/epoch - 5ms/step Epoch 11/500 Epoch 00011: val_loss improved from 0.00390 to 0.00352, saving model to LSTM6.h5 58/58 - 0s - loss: 0.0050 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 0.0010 - 302ms/epoch - 5ms/step Epoch 12/500 Epoch 00012: val_loss improved from 0.00352 to 0.00301, saving model to LSTM6.h5 58/58 - 0s - loss: 0.0028 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 0.0010 - 299ms/epoch - 5ms/step Epoch 13/500 Epoch 00013: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0025 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 0.0010 - 272ms/epoch - 5ms/step Epoch 14/500 Epoch 00014: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0030 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 0.0010 - 270ms/epoch - 5ms/step Epoch 15/500 Epoch 00015: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0041 - accuracy: 0.0000e+00 - val_loss: 0.0040 - val_accuracy: 0.0037 - lr: 0.0010 - 265ms/epoch - 5ms/step Epoch 16/500 Epoch 00016: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0068 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 0.0010 - 268ms/epoch - 5ms/step Epoch 17/500 Epoch 00017: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513. Epoch 00017: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0117 - accuracy: 0.0000e+00 - val_loss: 0.0053 - val_accuracy: 0.0037 - lr: 0.0010 - 276ms/epoch - 5ms/step Epoch 18/500 Epoch 00018: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0179 - accuracy: 0.0000e+00 - val_loss: 0.0140 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 290ms/epoch - 5ms/step Epoch 19/500 Epoch 00019: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0028 - accuracy: 0.0000e+00 - val_loss: 0.0090 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 297ms/epoch - 5ms/step Epoch 20/500 Epoch 00020: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0020 - accuracy: 0.0000e+00 - val_loss: 0.0075 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 262ms/epoch - 5ms/step Epoch 21/500 Epoch 00021: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0015 - accuracy: 0.0000e+00 - val_loss: 0.0061 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 294ms/epoch - 5ms/step Epoch 22/500 Epoch 00022: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05. Epoch 00022: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 292ms/epoch - 5ms/step Epoch 23/500 Epoch 00023: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 293ms/epoch - 5ms/step Epoch 24/500 Epoch 00024: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step Epoch 25/500 Epoch 00025: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0049 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 282ms/epoch - 5ms/step Epoch 26/500 Epoch 00026: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0048 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 276ms/epoch - 5ms/step Epoch 27/500 Epoch 00027: ReduceLROnPlateau reducing learning rate to 1e-05. Epoch 00027: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0047 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 274ms/epoch - 5ms/step Epoch 28/500 Epoch 00028: val_loss did not improve from 0.00301 58/58 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 272ms/epoch - 5ms/step Epoch 29/500 Epoch 00029: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.9450e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 288ms/epoch - 5ms/step Epoch 30/500 Epoch 00030: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.8684e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 285ms/epoch - 5ms/step Epoch 31/500 Epoch 00031: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.7942e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 272ms/epoch - 5ms/step Epoch 32/500 Epoch 00032: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.7223e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 291ms/epoch - 5ms/step Epoch 33/500 Epoch 00033: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.6527e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 275ms/epoch - 5ms/step Epoch 34/500 Epoch 00034: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.5853e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 279ms/epoch - 5ms/step Epoch 35/500 Epoch 00035: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.5201e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step Epoch 36/500 Epoch 00036: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.4571e-04 - accuracy: 0.0000e+00 - val_loss: 0.0040 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 264ms/epoch - 5ms/step Epoch 37/500 Epoch 00037: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.3964e-04 - accuracy: 0.0000e+00 - val_loss: 0.0039 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 295ms/epoch - 5ms/step Epoch 38/500 Epoch 00038: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.3379e-04 - accuracy: 0.0000e+00 - val_loss: 0.0038 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 278ms/epoch - 5ms/step Epoch 39/500 Epoch 00039: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.2816e-04 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 277ms/epoch - 5ms/step Epoch 40/500 Epoch 00040: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.2275e-04 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 264ms/epoch - 5ms/step Epoch 41/500 Epoch 00041: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.1755e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 278ms/epoch - 5ms/step Epoch 42/500 Epoch 00042: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.1257e-04 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 269ms/epoch - 5ms/step Epoch 43/500 Epoch 00043: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.0779e-04 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step Epoch 44/500 Epoch 00044: val_loss did not improve from 0.00301 58/58 - 0s - loss: 9.0322e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 283ms/epoch - 5ms/step Epoch 45/500 Epoch 00045: val_loss did not improve from 0.00301 58/58 - 0s - loss: 8.9885e-04 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 268ms/epoch - 5ms/step Epoch 46/500 Epoch 00046: val_loss did not improve from 0.00301 58/58 - 0s - loss: 8.9466e-04 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 276ms/epoch - 5ms/step Epoch 47/500 Epoch 00047: val_loss did not improve from 0.00301 58/58 - 0s - loss: 8.9066e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 283ms/epoch - 5ms/step Epoch 48/500 Epoch 00048: val_loss did not improve from 0.00301 58/58 - 0s - loss: 8.8683e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step Epoch 49/500 Epoch 00049: val_loss did not improve from 0.00301 58/58 - 0s - loss: 8.8318e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 272ms/epoch - 5ms/step Epoch 50/500 Epoch 00050: val_loss did not improve from 0.00301 58/58 - 0s - loss: 8.7968e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 277ms/epoch - 5ms/step Epoch 51/500 Epoch 00051: val_loss improved from 0.00301 to 0.00300, saving model to LSTM6.h5 58/58 - 1s - loss: 8.7634e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 509ms/epoch - 9ms/step Epoch 52/500 Epoch 00052: val_loss improved from 0.00300 to 0.00295, saving model to LSTM6.h5 58/58 - 0s - loss: 8.7314e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 286ms/epoch - 5ms/step Epoch 53/500 Epoch 00053: val_loss improved from 0.00295 to 0.00291, saving model to LSTM6.h5 58/58 - 0s - loss: 8.7008e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 281ms/epoch - 5ms/step Epoch 54/500 Epoch 00054: val_loss improved from 0.00291 to 0.00287, saving model to LSTM6.h5 58/58 - 0s - loss: 8.6714e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 279ms/epoch - 5ms/step Epoch 55/500 Epoch 00055: val_loss improved from 0.00287 to 0.00283, saving model to LSTM6.h5 58/58 - 0s - loss: 8.6433e-04 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 316ms/epoch - 5ms/step Epoch 56/500 Epoch 00056: val_loss improved from 0.00283 to 0.00279, saving model to LSTM6.h5 58/58 - 0s - loss: 8.6163e-04 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 286ms/epoch - 5ms/step Epoch 57/500 Epoch 00057: val_loss improved from 0.00279 to 0.00275, saving model to LSTM6.h5 58/58 - 0s - loss: 8.5903e-04 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 278ms/epoch - 5ms/step Epoch 58/500 Epoch 00058: val_loss improved from 0.00275 to 0.00272, saving model to LSTM6.h5 58/58 - 0s - loss: 8.5653e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 283ms/epoch - 5ms/step Epoch 59/500 Epoch 00059: val_loss improved from 0.00272 to 0.00269, saving model to LSTM6.h5 58/58 - 0s - loss: 8.5413e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 284ms/epoch - 5ms/step Epoch 60/500 Epoch 00060: val_loss improved from 0.00269 to 0.00266, saving model to LSTM6.h5 58/58 - 0s - loss: 8.5180e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 292ms/epoch - 5ms/step Epoch 61/500 Epoch 00061: val_loss improved from 0.00266 to 0.00264, saving model to LSTM6.h5 58/58 - 0s - loss: 8.4955e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 282ms/epoch - 5ms/step Epoch 62/500 Epoch 00062: val_loss improved from 0.00264 to 0.00262, saving model to LSTM6.h5 58/58 - 0s - loss: 8.4737e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 292ms/epoch - 5ms/step Epoch 63/500 Epoch 00063: val_loss improved from 0.00262 to 0.00259, saving model to LSTM6.h5 58/58 - 0s - loss: 8.4525e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 279ms/epoch - 5ms/step Epoch 64/500 Epoch 00064: val_loss improved from 0.00259 to 0.00258, saving model to LSTM6.h5 58/58 - 0s - loss: 8.4319e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 285ms/epoch - 5ms/step Epoch 65/500 Epoch 00065: val_loss improved from 0.00258 to 0.00256, saving model to LSTM6.h5 58/58 - 0s - loss: 8.4117e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 288ms/epoch - 5ms/step Epoch 66/500 Epoch 00066: val_loss improved from 0.00256 to 0.00255, saving model to LSTM6.h5 58/58 - 0s - loss: 8.3921e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 276ms/epoch - 5ms/step Epoch 67/500 Epoch 00067: val_loss improved from 0.00255 to 0.00253, saving model to LSTM6.h5 58/58 - 0s - loss: 8.3728e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 273ms/epoch - 5ms/step Epoch 68/500 Epoch 00068: val_loss improved from 0.00253 to 0.00252, saving model to LSTM6.h5 58/58 - 0s - loss: 8.3539e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 277ms/epoch - 5ms/step Epoch 69/500 Epoch 00069: val_loss improved from 0.00252 to 0.00251, saving model to LSTM6.h5 58/58 - 0s - loss: 8.3352e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 289ms/epoch - 5ms/step Epoch 70/500 Epoch 00070: val_loss improved from 0.00251 to 0.00251, saving model to LSTM6.h5 58/58 - 0s - loss: 8.3168e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 274ms/epoch - 5ms/step Epoch 71/500 Epoch 00071: val_loss improved from 0.00251 to 0.00250, saving model to LSTM6.h5 58/58 - 0s - loss: 8.2987e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 273ms/epoch - 5ms/step Epoch 72/500 Epoch 00072: val_loss improved from 0.00250 to 0.00250, saving model to LSTM6.h5 58/58 - 0s - loss: 8.2807e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 284ms/epoch - 5ms/step Epoch 73/500 Epoch 00073: val_loss improved from 0.00250 to 0.00250, saving model to LSTM6.h5 58/58 - 0s - loss: 8.2628e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 282ms/epoch - 5ms/step Epoch 74/500 Epoch 00074: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.2451e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step Epoch 75/500 Epoch 00075: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.2274e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 275ms/epoch - 5ms/step Epoch 76/500 Epoch 00076: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.2098e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 276ms/epoch - 5ms/step Epoch 77/500 Epoch 00077: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.1922e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 272ms/epoch - 5ms/step Epoch 78/500 Epoch 00078: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.1746e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 254ms/epoch - 4ms/step Epoch 79/500 Epoch 00079: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.1570e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 260ms/epoch - 4ms/step Epoch 80/500 Epoch 00080: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.1394e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 274ms/epoch - 5ms/step Epoch 81/500 Epoch 00081: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.1217e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 281ms/epoch - 5ms/step Epoch 82/500 Epoch 00082: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.1039e-04 - accuracy: 0.0000e+00 - val_loss: 0.0025 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 261ms/epoch - 4ms/step Epoch 83/500 Epoch 00083: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.0861e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 285ms/epoch - 5ms/step Epoch 84/500 Epoch 00084: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.0682e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 267ms/epoch - 5ms/step Epoch 85/500 Epoch 00085: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.0502e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 293ms/epoch - 5ms/step Epoch 86/500 Epoch 00086: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.0320e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step Epoch 87/500 Epoch 00087: val_loss did not improve from 0.00250 58/58 - 0s - loss: 8.0138e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 282ms/epoch - 5ms/step Epoch 88/500 Epoch 00088: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.9955e-04 - accuracy: 0.0000e+00 - val_loss: 0.0026 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 263ms/epoch - 5ms/step Epoch 89/500 Epoch 00089: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.9770e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 267ms/epoch - 5ms/step Epoch 90/500 Epoch 00090: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.9585e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step Epoch 91/500 Epoch 00091: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.9398e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 262ms/epoch - 5ms/step Epoch 92/500 Epoch 00092: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.9209e-04 - accuracy: 0.0000e+00 - val_loss: 0.0027 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 260ms/epoch - 4ms/step Epoch 93/500 Epoch 00093: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.9020e-04 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 280ms/epoch - 5ms/step Epoch 94/500 Epoch 00094: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.8829e-04 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step Epoch 95/500 Epoch 00095: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.8637e-04 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 268ms/epoch - 5ms/step Epoch 96/500 Epoch 00096: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.8444e-04 - accuracy: 0.0000e+00 - val_loss: 0.0028 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 255ms/epoch - 4ms/step Epoch 97/500 Epoch 00097: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.8250e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 273ms/epoch - 5ms/step Epoch 98/500 Epoch 00098: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.8055e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step Epoch 99/500 Epoch 00099: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.7858e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 267ms/epoch - 5ms/step Epoch 100/500 Epoch 00100: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.7660e-04 - accuracy: 0.0000e+00 - val_loss: 0.0029 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step Epoch 101/500 Epoch 00101: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.7462e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step Epoch 102/500 Epoch 00102: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.7262e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step Epoch 103/500 Epoch 00103: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.7061e-04 - accuracy: 0.0000e+00 - val_loss: 0.0030 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 262ms/epoch - 5ms/step Epoch 104/500 Epoch 00104: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.6859e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 261ms/epoch - 4ms/step Epoch 105/500 Epoch 00105: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.6657e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 281ms/epoch - 5ms/step Epoch 106/500 Epoch 00106: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.6454e-04 - accuracy: 0.0000e+00 - val_loss: 0.0031 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 274ms/epoch - 5ms/step Epoch 107/500 Epoch 00107: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.6249e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 281ms/epoch - 5ms/step Epoch 108/500 Epoch 00108: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.6045e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 262ms/epoch - 5ms/step Epoch 109/500 Epoch 00109: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.5839e-04 - accuracy: 0.0000e+00 - val_loss: 0.0032 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step Epoch 110/500 Epoch 00110: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.5633e-04 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 263ms/epoch - 5ms/step Epoch 111/500 Epoch 00111: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.5427e-04 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 261ms/epoch - 5ms/step Epoch 112/500 Epoch 00112: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.5220e-04 - accuracy: 0.0000e+00 - val_loss: 0.0033 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 262ms/epoch - 5ms/step Epoch 113/500 Epoch 00113: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.5012e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step Epoch 114/500 Epoch 00114: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.4805e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step Epoch 115/500 Epoch 00115: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.4597e-04 - accuracy: 0.0000e+00 - val_loss: 0.0034 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 263ms/epoch - 5ms/step Epoch 116/500 Epoch 00116: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.4389e-04 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step Epoch 117/500 Epoch 00117: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.4181e-04 - accuracy: 0.0000e+00 - val_loss: 0.0035 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 275ms/epoch - 5ms/step Epoch 118/500 Epoch 00118: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.3973e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step Epoch 119/500 Epoch 00119: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.3765e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 265ms/epoch - 5ms/step Epoch 120/500 Epoch 00120: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.3557e-04 - accuracy: 0.0000e+00 - val_loss: 0.0036 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 259ms/epoch - 4ms/step Epoch 121/500 Epoch 00121: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.3349e-04 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 281ms/epoch - 5ms/step Epoch 122/500 Epoch 00122: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.3141e-04 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 278ms/epoch - 5ms/step Epoch 123/500 Epoch 00123: val_loss did not improve from 0.00250 58/58 - 0s - loss: 7.2934e-04 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 257ms/epoch - 4ms/step Epoch 00123: early stopping
SMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 71.66891031373025
RMSE: 8.465749247038342
MAPE: 6.880610177712922
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 67.24711347278334
RMSE: 8.200433736869249
MAPE: 6.781803215137433
WMA
Prediction vs Close: 55.6% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 58.550384022767716
RMSE: 7.6518222681115455
MAPE: 6.1413991074844
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 127.72881036471918
RMSE: 11.301717142307146
MAPE: 10.306940424406019
KAMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 48.51% Accuracy
MSE: 43.855468679122
RMSE: 6.622346161227303
MAPE: 5.4751276749367985
MIDPOINT
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 45.15% Accuracy
MSE: 67.44416314351042
RMSE: 8.212439536673035
MAPE: 6.768235104271493
T3
T3([input_arrays], [timeperiod=5], [vfactor=0.7])
Triple Exponential Moving Average (T3) (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 5
vfactor: 0.7
Outputs:
real
19
Working on T3 predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-16569.270, Time=2.21 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14511.291, Time=2.35 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15408.738, Time=7.62 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-15165.005, Time=8.22 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15595.465, Time=6.91 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15837.470, Time=10.11 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-15491.538, Time=9.25 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-16378.438, Time=2.50 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-16318.604, Time=3.49 sec
ARIMA(1,3,1)(0,0,0)[0] intercept : AIC=-16567.270, Time=2.32 sec
Best model: ARIMA(1,3,1)(0,0,0)[0]
Total fit time: 54.991 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(1, 3, 1) Log Likelihood 8308.635
Date: Sun, 12 Dec 2021 AIC -16569.270
Time: 17:10:07 BIC -16456.690
Sample: 0 HQIC -16526.035
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 1.355e-13 3.43e-05 3.95e-09 1.000 -6.72e-05 6.72e-05
x2 5.009e-14 2.67e-05 1.88e-09 1.000 -5.23e-05 5.23e-05
x3 -9.101e-15 2.19e-05 -4.16e-10 1.000 -4.29e-05 4.29e-05
x4 1.0000 2.97e-05 3.37e+04 0.000 1.000 1.000
x5 3.626e-12 3.2e-05 1.13e-07 1.000 -6.28e-05 6.28e-05
x6 6.879e-17 0.000 5.13e-13 1.000 -0.000 0.000
x7 1.588e-13 4.04e-05 3.93e-09 1.000 -7.92e-05 7.92e-05
x8 -0.0002 9.77e-06 -20.395 0.000 -0.000 -0.000
x9 3.877e-14 0.001 6.24e-11 1.000 -0.001 0.001
x10 -7.41e-05 0.001 -0.129 0.897 -0.001 0.001
x11 0.0003 4.91e-05 6.926 0.000 0.000 0.000
x12 -0.0004 7.27e-05 -5.556 0.000 -0.001 -0.000
x13 -2.679e-14 3.39e-05 -7.9e-10 1.000 -6.65e-05 6.65e-05
x14 2.97e-13 0.000 2.31e-09 1.000 -0.000 0.000
x15 1.602e-12 7.47e-05 2.14e-08 1.000 -0.000 0.000
x16 -8.756e-13 4.29e-05 -2.04e-08 1.000 -8.41e-05 8.41e-05
x17 1.793e-12 6.56e-05 2.74e-08 1.000 -0.000 0.000
x18 -1.019e-13 0.000 -5.54e-10 1.000 -0.000 0.000
x19 -1.077e-12 8.29e-05 -1.3e-08 1.000 -0.000 0.000
x20 1.771e-13 8.45e-05 2.1e-09 1.000 -0.000 0.000
x21 9.233e-16 0.000 1.94e-12 1.000 -0.001 0.001
ar.L1 -0.2857 0.000 -2747.572 0.000 -0.286 -0.285
ma.L1 -0.9142 7.12e-06 -1.28e+05 0.000 -0.914 -0.914
sigma2 1e-10 7e-11 1.429 0.153 -3.71e-11 2.37e-10
===================================================================================
Ljung-Box (L1) (Q): 84.32 Jarque-Bera (JB): 4804295.53
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.87
Prob(H) (two-sided): 0.00 Kurtosis: 381.28
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 2.06e+20. Standard errors may be unstable.
ARIMA order: (1, 3, 1)
/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead. super(Adam, self).__init__(name, **kwargs)
Epoch 1/500 Epoch 00001: val_loss improved from inf to 0.03167, saving model to LSTM6.h5 43/43 - 4s - loss: 0.1223 - accuracy: 0.0000e+00 - val_loss: 0.0317 - val_accuracy: 0.0037 - lr: 0.0010 - 4s/epoch - 97ms/step Epoch 2/500 Epoch 00002: val_loss did not improve from 0.03167 43/43 - 0s - loss: 0.0537 - accuracy: 0.0000e+00 - val_loss: 0.0551 - val_accuracy: 0.0037 - lr: 0.0010 - 213ms/epoch - 5ms/step Epoch 3/500 Epoch 00003: val_loss did not improve from 0.03167 43/43 - 0s - loss: 0.0270 - accuracy: 0.0000e+00 - val_loss: 0.0575 - val_accuracy: 0.0037 - lr: 0.0010 - 208ms/epoch - 5ms/step Epoch 4/500 Epoch 00004: val_loss improved from 0.03167 to 0.01264, saving model to LSTM6.h5 43/43 - 0s - loss: 0.0175 - accuracy: 0.0000e+00 - val_loss: 0.0126 - val_accuracy: 0.0037 - lr: 0.0010 - 222ms/epoch - 5ms/step Epoch 5/500 Epoch 00005: val_loss did not improve from 0.01264 43/43 - 0s - loss: 0.0093 - accuracy: 0.0000e+00 - val_loss: 0.0307 - val_accuracy: 0.0037 - lr: 0.0010 - 214ms/epoch - 5ms/step Epoch 6/500 Epoch 00006: val_loss improved from 0.01264 to 0.00368, saving model to LSTM6.h5 43/43 - 0s - loss: 0.0060 - accuracy: 0.0000e+00 - val_loss: 0.0037 - val_accuracy: 0.0037 - lr: 0.0010 - 242ms/epoch - 6ms/step Epoch 7/500 Epoch 00007: val_loss did not improve from 0.00368 43/43 - 0s - loss: 0.0029 - accuracy: 0.0000e+00 - val_loss: 0.0188 - val_accuracy: 0.0037 - lr: 0.0010 - 221ms/epoch - 5ms/step Epoch 8/500 Epoch 00008: val_loss did not improve from 0.00368 43/43 - 0s - loss: 0.0021 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 0.0010 - 218ms/epoch - 5ms/step Epoch 9/500 Epoch 00009: val_loss did not improve from 0.00368 43/43 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0129 - val_accuracy: 0.0037 - lr: 0.0010 - 217ms/epoch - 5ms/step Epoch 10/500 Epoch 00010: val_loss did not improve from 0.00368 43/43 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0057 - val_accuracy: 0.0037 - lr: 0.0010 - 225ms/epoch - 5ms/step Epoch 11/500 Epoch 00011: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513. Epoch 00011: val_loss did not improve from 0.00368 43/43 - 0s - loss: 9.6307e-04 - accuracy: 0.0000e+00 - val_loss: 0.0099 - val_accuracy: 0.0037 - lr: 0.0010 - 217ms/epoch - 5ms/step Epoch 12/500 Epoch 00012: val_loss did not improve from 0.00368 43/43 - 0s - loss: 0.0016 - accuracy: 0.0000e+00 - val_loss: 0.0056 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 204ms/epoch - 5ms/step Epoch 13/500 Epoch 00013: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.9181e-04 - accuracy: 0.0000e+00 - val_loss: 0.0055 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 196ms/epoch - 5ms/step Epoch 14/500 Epoch 00014: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.8964e-04 - accuracy: 0.0000e+00 - val_loss: 0.0050 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 207ms/epoch - 5ms/step Epoch 15/500 Epoch 00015: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.6453e-04 - accuracy: 0.0000e+00 - val_loss: 0.0048 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 235ms/epoch - 5ms/step Epoch 16/500 Epoch 00016: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05. Epoch 00016: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.5108e-04 - accuracy: 0.0000e+00 - val_loss: 0.0046 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 222ms/epoch - 5ms/step Epoch 17/500 Epoch 00017: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.2235e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 204ms/epoch - 5ms/step Epoch 18/500 Epoch 00018: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.1948e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 198ms/epoch - 5ms/step Epoch 19/500 Epoch 00019: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.1750e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 205ms/epoch - 5ms/step Epoch 20/500 Epoch 00020: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.1615e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step Epoch 21/500 Epoch 00021: ReduceLROnPlateau reducing learning rate to 1e-05. Epoch 00021: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.1509e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 212ms/epoch - 5ms/step Epoch 22/500 Epoch 00022: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.1416e-04 - accuracy: 0.0000e+00 - val_loss: 0.0045 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step Epoch 23/500 Epoch 00023: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.1329e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step Epoch 24/500 Epoch 00024: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.1243e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step Epoch 25/500 Epoch 00025: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.1158e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step Epoch 26/500 Epoch 00026: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.1073e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step Epoch 27/500 Epoch 00027: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0986e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 197ms/epoch - 5ms/step Epoch 28/500 Epoch 00028: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0898e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step Epoch 29/500 Epoch 00029: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0808e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 228ms/epoch - 5ms/step Epoch 30/500 Epoch 00030: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0717e-04 - accuracy: 0.0000e+00 - val_loss: 0.0044 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 211ms/epoch - 5ms/step Epoch 31/500 Epoch 00031: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0625e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 202ms/epoch - 5ms/step Epoch 32/500 Epoch 00032: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0531e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 198ms/epoch - 5ms/step Epoch 33/500 Epoch 00033: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0436e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step Epoch 34/500 Epoch 00034: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0339e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step Epoch 35/500 Epoch 00035: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0241e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step Epoch 36/500 Epoch 00036: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0142e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 214ms/epoch - 5ms/step Epoch 37/500 Epoch 00037: val_loss did not improve from 0.00368 43/43 - 0s - loss: 8.0041e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 212ms/epoch - 5ms/step Epoch 38/500 Epoch 00038: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.9940e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 197ms/epoch - 5ms/step Epoch 39/500 Epoch 00039: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.9837e-04 - accuracy: 0.0000e+00 - val_loss: 0.0043 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 215ms/epoch - 5ms/step Epoch 40/500 Epoch 00040: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.9733e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 204ms/epoch - 5ms/step Epoch 41/500 Epoch 00041: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.9628e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 211ms/epoch - 5ms/step Epoch 42/500 Epoch 00042: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.9523e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step Epoch 43/500 Epoch 00043: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.9416e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 229ms/epoch - 5ms/step Epoch 44/500 Epoch 00044: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.9308e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step Epoch 45/500 Epoch 00045: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.9200e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 211ms/epoch - 5ms/step Epoch 46/500 Epoch 00046: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.9091e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 199ms/epoch - 5ms/step Epoch 47/500 Epoch 00047: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.8981e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step Epoch 48/500 Epoch 00048: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.8870e-04 - accuracy: 0.0000e+00 - val_loss: 0.0042 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step Epoch 49/500 Epoch 00049: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.8759e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step Epoch 50/500 Epoch 00050: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.8647e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 198ms/epoch - 5ms/step Epoch 51/500 Epoch 00051: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.8535e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 211ms/epoch - 5ms/step Epoch 52/500 Epoch 00052: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.8422e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step Epoch 53/500 Epoch 00053: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.8309e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step Epoch 54/500 Epoch 00054: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.8195e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 214ms/epoch - 5ms/step Epoch 55/500 Epoch 00055: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.8081e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step Epoch 56/500 Epoch 00056: val_loss did not improve from 0.00368 43/43 - 0s - loss: 7.7966e-04 - accuracy: 0.0000e+00 - val_loss: 0.0041 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step Epoch 00056: early stopping
SMA
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 71.66891031373025
RMSE: 8.465749247038342
MAPE: 6.880610177712922
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 43.66% Accuracy
MSE: 67.24711347278334
RMSE: 8.200433736869249
MAPE: 6.781803215137433
WMA
Prediction vs Close: 55.6% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 58.550384022767716
RMSE: 7.6518222681115455
MAPE: 6.1413991074844
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 127.72881036471918
RMSE: 11.301717142307146
MAPE: 10.306940424406019
KAMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 48.51% Accuracy
MSE: 43.855468679122
RMSE: 6.622346161227303
MAPE: 5.4751276749367985
MIDPOINT
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 45.15% Accuracy
MSE: 67.44416314351042
RMSE: 8.212439536673035
MAPE: 6.768235104271493
T3
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 44.78% Accuracy
MSE: 154.873959597027
RMSE: 12.444836664136135
MAPE: 10.329006454112236
TEMA
TEMA([input_arrays], [timeperiod=30])
Triple Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
9
Working on TEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-16493.570, Time=2.72 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-15527.581, Time=7.39 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16154.477, Time=7.18 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-15134.948, Time=6.74 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-16538.454, Time=8.49 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-16271.346, Time=2.19 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-16350.992, Time=13.04 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-16200.149, Time=3.34 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-16461.809, Time=16.02 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-16384.147, Time=3.44 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=9.34 sec
ARIMA(2,3,1)(0,0,0)[0] intercept : AIC=-15110.164, Time=5.45 sec
Best model: ARIMA(2,3,1)(0,0,0)[0]
Total fit time: 85.353 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(2, 3, 1) Log Likelihood 8294.227
Date: Sun, 12 Dec 2021 AIC -16538.454
Time: 17:14:34 BIC -16421.183
Sample: 0 HQIC -16493.417
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 3.591e-07 0.001 0.000 1.000 -0.002 0.002
x2 3.6e-07 0.002 0.000 1.000 -0.003 0.003
x3 3.611e-07 0.001 0.000 1.000 -0.002 0.002
x4 1.0000 0.000 2628.605 0.000 0.999 1.001
x5 3.432e-07 0.000 0.001 0.999 -0.001 0.001
x6 1.714e-07 4.05e-05 0.004 0.997 -7.91e-05 7.95e-05
x7 3.541e-07 0.001 0.000 1.000 -0.003 0.003
x8 -0.0002 0.000 -1.006 0.315 -0.001 0.000
x9 -7.559e-08 0.000 -0.000 1.000 -0.001 0.001
x10 0.0001 0.000 0.492 0.623 -0.000 0.001
x11 -0.0006 0.000 -2.697 0.007 -0.001 -0.000
x12 0.0005 0.000 1.741 0.082 -5.97e-05 0.001
x13 3.6e-07 0.000 0.002 0.999 -0.000 0.000
x14 1.003e-06 0.001 0.001 0.999 -0.002 0.002
x15 3.506e-07 7.16e-05 0.005 0.996 -0.000 0.000
x16 5.157e-07 0.000 0.005 0.996 -0.000 0.000
x17 3.516e-07 6.59e-05 0.005 0.996 -0.000 0.000
x18 1.166e-07 0.000 0.001 1.000 -0.000 0.000
x19 3.922e-07 7.5e-05 0.005 0.996 -0.000 0.000
x20 -3.64e-07 0.000 -0.002 0.999 -0.000 0.000
x21 4.458e-07 0.000 0.004 0.997 -0.000 0.000
ar.L1 -0.4019 4.12e-05 -9758.484 0.000 -0.402 -0.402
ar.L2 -0.1006 1.58e-05 -6360.873 0.000 -0.101 -0.101
ma.L1 -0.7963 8.45e-06 -9.43e+04 0.000 -0.796 -0.796
sigma2 9.048e-11 7.2e-11 1.257 0.209 -5.06e-11 2.32e-10
===================================================================================
Ljung-Box (L1) (Q): 64.02 Jarque-Bera (JB): 4424775.47
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.53
Prob(H) (two-sided): 0.00 Kurtosis: 366.04
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.02e+20. Standard errors may be unstable.
ARIMA order: (2, 3, 1)
/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead. super(Adam, self).__init__(name, **kwargs)
Epoch 1/500 Epoch 00001: val_loss improved from inf to 0.08025, saving model to LSTM6.h5 90/90 - 4s - loss: 0.1296 - accuracy: 0.0000e+00 - val_loss: 0.0802 - val_accuracy: 0.0037 - lr: 0.0010 - 4s/epoch - 46ms/step Epoch 2/500 Epoch 00002: val_loss improved from 0.08025 to 0.00870, saving model to LSTM6.h5 90/90 - 0s - loss: 0.0263 - accuracy: 0.0000e+00 - val_loss: 0.0087 - val_accuracy: 0.0037 - lr: 0.0010 - 446ms/epoch - 5ms/step Epoch 3/500 Epoch 00003: val_loss improved from 0.00870 to 0.00668, saving model to LSTM6.h5 90/90 - 0s - loss: 0.0346 - accuracy: 0.0000e+00 - val_loss: 0.0067 - val_accuracy: 0.0037 - lr: 0.0010 - 427ms/epoch - 5ms/step Epoch 4/500 Epoch 00004: val_loss improved from 0.00668 to 0.00660, saving model to LSTM6.h5 90/90 - 0s - loss: 0.0286 - accuracy: 0.0000e+00 - val_loss: 0.0066 - val_accuracy: 0.0037 - lr: 0.0010 - 437ms/epoch - 5ms/step Epoch 5/500 Epoch 00005: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0280 - accuracy: 0.0000e+00 - val_loss: 0.0074 - val_accuracy: 0.0037 - lr: 0.0010 - 419ms/epoch - 5ms/step Epoch 6/500 Epoch 00006: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0229 - accuracy: 0.0000e+00 - val_loss: 0.0081 - val_accuracy: 0.0037 - lr: 0.0010 - 417ms/epoch - 5ms/step Epoch 7/500 Epoch 00007: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0201 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 0.0010 - 424ms/epoch - 5ms/step Epoch 8/500 Epoch 00008: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513. Epoch 00008: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0167 - accuracy: 0.0000e+00 - val_loss: 0.0078 - val_accuracy: 0.0037 - lr: 0.0010 - 420ms/epoch - 5ms/step Epoch 9/500 Epoch 00009: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0206 - accuracy: 0.0000e+00 - val_loss: 0.0484 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 435ms/epoch - 5ms/step Epoch 10/500 Epoch 00010: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0047 - accuracy: 0.0000e+00 - val_loss: 0.0377 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 399ms/epoch - 4ms/step Epoch 11/500 Epoch 00011: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0034 - accuracy: 0.0000e+00 - val_loss: 0.0311 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 427ms/epoch - 5ms/step Epoch 12/500 Epoch 00012: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0024 - accuracy: 0.0000e+00 - val_loss: 0.0263 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 423ms/epoch - 5ms/step Epoch 13/500 Epoch 00013: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05. Epoch 00013: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0018 - accuracy: 0.0000e+00 - val_loss: 0.0223 - val_accuracy: 0.0037 - lr: 1.0000e-04 - 404ms/epoch - 4ms/step Epoch 14/500 Epoch 00014: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0015 - accuracy: 0.0000e+00 - val_loss: 0.0227 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 430ms/epoch - 5ms/step Epoch 15/500 Epoch 00015: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0228 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 408ms/epoch - 5ms/step Epoch 16/500 Epoch 00016: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0014 - accuracy: 0.0000e+00 - val_loss: 0.0228 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 422ms/epoch - 5ms/step Epoch 17/500 Epoch 00017: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0227 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 445ms/epoch - 5ms/step Epoch 18/500 Epoch 00018: ReduceLROnPlateau reducing learning rate to 1e-05. Epoch 00018: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0224 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 403ms/epoch - 4ms/step Epoch 19/500 Epoch 00019: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0221 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 416ms/epoch - 5ms/step Epoch 20/500 Epoch 00020: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0013 - accuracy: 0.0000e+00 - val_loss: 0.0217 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 403ms/epoch - 4ms/step Epoch 21/500 Epoch 00021: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0213 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 419ms/epoch - 5ms/step Epoch 22/500 Epoch 00022: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0208 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 410ms/epoch - 5ms/step Epoch 23/500 Epoch 00023: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0204 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 412ms/epoch - 5ms/step Epoch 24/500 Epoch 00024: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0199 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 425ms/epoch - 5ms/step Epoch 25/500 Epoch 00025: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0012 - accuracy: 0.0000e+00 - val_loss: 0.0194 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 424ms/epoch - 5ms/step Epoch 26/500 Epoch 00026: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0189 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 410ms/epoch - 5ms/step Epoch 27/500 Epoch 00027: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0185 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 413ms/epoch - 5ms/step Epoch 28/500 Epoch 00028: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0180 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 453ms/epoch - 5ms/step Epoch 29/500 Epoch 00029: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0175 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 396ms/epoch - 4ms/step Epoch 30/500 Epoch 00030: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0170 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 408ms/epoch - 5ms/step Epoch 31/500 Epoch 00031: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0165 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 414ms/epoch - 5ms/step Epoch 32/500 Epoch 00032: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0011 - accuracy: 0.0000e+00 - val_loss: 0.0160 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 400ms/epoch - 4ms/step Epoch 33/500 Epoch 00033: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0155 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 445ms/epoch - 5ms/step Epoch 34/500 Epoch 00034: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0150 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 417ms/epoch - 5ms/step Epoch 35/500 Epoch 00035: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0145 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 447ms/epoch - 5ms/step Epoch 36/500 Epoch 00036: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0140 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 430ms/epoch - 5ms/step Epoch 37/500 Epoch 00037: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0135 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 417ms/epoch - 5ms/step Epoch 38/500 Epoch 00038: val_loss did not improve from 0.00660 90/90 - 0s - loss: 0.0010 - accuracy: 0.0000e+00 - val_loss: 0.0130 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 426ms/epoch - 5ms/step Epoch 39/500 Epoch 00039: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.9877e-04 - accuracy: 0.0000e+00 - val_loss: 0.0125 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 418ms/epoch - 5ms/step Epoch 40/500 Epoch 00040: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.9247e-04 - accuracy: 0.0000e+00 - val_loss: 0.0121 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 448ms/epoch - 5ms/step Epoch 41/500 Epoch 00041: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.8666e-04 - accuracy: 0.0000e+00 - val_loss: 0.0116 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 440ms/epoch - 5ms/step Epoch 42/500 Epoch 00042: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.8126e-04 - accuracy: 0.0000e+00 - val_loss: 0.0112 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 438ms/epoch - 5ms/step Epoch 43/500 Epoch 00043: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.7624e-04 - accuracy: 0.0000e+00 - val_loss: 0.0107 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 428ms/epoch - 5ms/step Epoch 44/500 Epoch 00044: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.7154e-04 - accuracy: 0.0000e+00 - val_loss: 0.0103 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 420ms/epoch - 5ms/step Epoch 45/500 Epoch 00045: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.6712e-04 - accuracy: 0.0000e+00 - val_loss: 0.0099 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 411ms/epoch - 5ms/step Epoch 46/500 Epoch 00046: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.6294e-04 - accuracy: 0.0000e+00 - val_loss: 0.0095 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 432ms/epoch - 5ms/step Epoch 47/500 Epoch 00047: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.5896e-04 - accuracy: 0.0000e+00 - val_loss: 0.0091 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 429ms/epoch - 5ms/step Epoch 48/500 Epoch 00048: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.5515e-04 - accuracy: 0.0000e+00 - val_loss: 0.0087 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 408ms/epoch - 5ms/step Epoch 49/500 Epoch 00049: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.5147e-04 - accuracy: 0.0000e+00 - val_loss: 0.0083 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 439ms/epoch - 5ms/step Epoch 50/500 Epoch 00050: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.4791e-04 - accuracy: 0.0000e+00 - val_loss: 0.0080 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 420ms/epoch - 5ms/step Epoch 51/500 Epoch 00051: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.4442e-04 - accuracy: 0.0000e+00 - val_loss: 0.0076 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 421ms/epoch - 5ms/step Epoch 52/500 Epoch 00052: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.4099e-04 - accuracy: 0.0000e+00 - val_loss: 0.0073 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 425ms/epoch - 5ms/step Epoch 53/500 Epoch 00053: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.3761e-04 - accuracy: 0.0000e+00 - val_loss: 0.0070 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 427ms/epoch - 5ms/step Epoch 54/500 Epoch 00054: val_loss did not improve from 0.00660 90/90 - 0s - loss: 9.3426e-04 - accuracy: 0.0000e+00 - val_loss: 0.0067 - val_accuracy: 0.0037 - lr: 1.0000e-05 - 444ms/epoch - 5ms/step Epoch 00054: early stopping
SMA Prediction vs Close: 53.73% Accuracy Prediction vs Prediction: 47.01% Accuracy MSE: 71.66891031373025 RMSE: 8.465749247038342 MAPE: 6.880610177712922 EMA Prediction vs Close: 54.85% Accuracy Prediction vs Prediction: 43.66% Accuracy MSE: 67.24711347278334 RMSE: 8.200433736869249 MAPE: 6.781803215137433 WMA Prediction vs Close: 55.6% Accuracy Prediction vs Prediction: 47.76% Accuracy MSE: 58.550384022767716 RMSE: 7.6518222681115455 MAPE: 6.1413991074844 DEMA Prediction vs Close: 52.24% Accuracy Prediction vs Prediction: 47.01% Accuracy MSE: 127.72881036471918 RMSE: 11.301717142307146 MAPE: 10.306940424406019 KAMA Prediction vs Close: 52.99% Accuracy Prediction vs Prediction: 48.51% Accuracy MSE: 43.855468679122 RMSE: 6.622346161227303 MAPE: 5.4751276749367985 MIDPOINT Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 45.15% Accuracy MSE: 67.44416314351042 RMSE: 8.212439536673035 MAPE: 6.768235104271493 T3 Prediction vs Close: 53.73% Accuracy Prediction vs Prediction: 44.78% Accuracy MSE: 154.873959597027 RMSE: 12.444836664136135 MAPE: 10.329006454112236 TEMA Prediction vs Close: 50.75% Accuracy Prediction vs Prediction: 47.01% Accuracy MSE: 163.5522149093863 RMSE: 12.788753454085597 MAPE: 11.455728191899736 Runtime: mins: 42.64405553298335
Architecture Used¶
In [ ]:
from google.colab import files
import cv2
uploaded = files.upload()
Saving Experiment6.png to Experiment6 (1).png
In [ ]:
imgfile = 'Experiment6'
img = cv2.imread('Experiment6.png')
plt.figure(figsize=(20,10))
plt.axis("off")
plt.title('LSTM Architecture '+imgfile,fontsize=18)
plt.imshow(img)
Out[ ]:
<matplotlib.image.AxesImage at 0x7fa52b8bb750>
Model Plots¶
In [81]:
with open('simulation6_data.json') as json_file:
simulation6 = json.load(json_file)
fileimg = 'Experiment6'
In [82]:
for i in range(len(list(simulation6.keys()))):
SIM = list(simulation6.keys())[i]
plot_train(simulation6,SIM)
plot_test(simulation6,SIM)
----- Train RMSE for SMA ----- 8.860010937586445 ----- Train_MSE_LSTM for SMA ----- 78.49979381415143 ----- Train MAE LSTM for SMA ----- 7.754632690589469
----- Test RMSE for SMA----- 8.465749247038342 ----- Test_MSE_LSTM for SMA----- 71.66891031373025 ----- Test_MAE_LSTM for SMA----- 6.880610177712922
----- Train RMSE for EMA ----- 10.185838832459739 ----- Train_MSE_LSTM for EMA ----- 103.7513127208448 ----- Train MAE LSTM for EMA ----- 9.021545289753416
----- Test RMSE for EMA----- 8.200433736869249 ----- Test_MSE_LSTM for EMA----- 67.24711347278334 ----- Test_MAE_LSTM for EMA----- 6.781803215137433
----- Train RMSE for WMA ----- 10.495797568969957 ----- Train_MSE_LSTM for WMA ----- 110.16176660879566 ----- Train MAE LSTM for WMA ----- 9.342000266276576
----- Test RMSE for WMA----- 7.6518222681115455 ----- Test_MSE_LSTM for WMA----- 58.550384022767716 ----- Test_MAE_LSTM for WMA----- 6.1413991074844
----- Train RMSE for DEMA ----- 12.196853963107756 ----- Train_MSE_LSTM for DEMA ----- 148.76324659737736 ----- Train MAE LSTM for DEMA ----- 10.958250443400392
----- Test RMSE for DEMA----- 11.301717142307146 ----- Test_MSE_LSTM for DEMA----- 127.72881036471918 ----- Test_MAE_LSTM for DEMA----- 10.306940424406019
----- Train RMSE for KAMA ----- 10.5711440318656 ----- Train_MSE_LSTM for KAMA ----- 111.74908614244768 ----- Train MAE LSTM for KAMA ----- 9.510206620484396
----- Test RMSE for KAMA----- 6.622346161227303 ----- Test_MSE_LSTM for KAMA----- 43.855468679122 ----- Test_MAE_LSTM for KAMA----- 5.4751276749367985
----- Train RMSE for MIDPOINT ----- 9.431256276275592 ----- Train_MSE_LSTM for MIDPOINT ----- 88.94859494878776 ----- Train MAE LSTM for MIDPOINT ----- 8.385353724073704
----- Test RMSE for MIDPOINT----- 8.212439536673035 ----- Test_MSE_LSTM for MIDPOINT----- 67.44416314351042 ----- Test_MAE_LSTM for MIDPOINT----- 6.768235104271493
----- Train RMSE for T3 ----- 12.0506886763047 ----- Train_MSE_LSTM for T3 ----- 145.21909757321833 ----- Train MAE LSTM for T3 ----- 10.839950464284371
----- Test RMSE for T3----- 12.444836664136135 ----- Test_MSE_LSTM for T3----- 154.873959597027 ----- Test_MAE_LSTM for T3----- 10.329006454112236
----- Train RMSE for TEMA ----- 7.410516533334238 ----- Train_MSE_LSTM for TEMA ----- 54.915755290820094 ----- Train MAE LSTM for TEMA ----- 5.083548934505515
----- Test RMSE for TEMA----- 12.788753454085597 ----- Test_MSE_LSTM for TEMA----- 163.5522149093863 ----- Test_MAE_LSTM for TEMA----- 11.455728191899736
Arima w Exogenous Variable Multistep MutiVariate LSTM Hybrid Model Experiment 7¶
In [ ]:
def get_arima_exog(dataframe,original_data, train_len, test_len):
# prepare train and test data for exogenous vr
X_value = pd.DataFrame(low_vol.iloc[:, :])
y_value = pd.DataFrame(low_vol.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
X_scale_dataset = X_scaler.fit_transform(X_value)
y_scale_dataset = y_scaler.fit_transform(y_value)
# Get data and check shape
# X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X_scale_dataset)
y_train, y_test, = split_train_test(y_scale_dataset)
yc_train,yc_test = split_train_test(low_vol_data)
yc = yc_test.values.tolist()
y_train_list = y_train.flatten().tolist()
y_test_list = y_test.flatten().tolist()
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
# Initialize model
model = auto_arima(y_train_list,exogenous = X_train,trace=True, error_action='ignore', start_p=1,start_q=1,max_p=3,max_q=3,d=3,
suppress_warnings=True,stepwise=True,seasonal=True)
# Determine model parameters
print(model.summary())
model.fit(y_train_list,maxiter=200)
order = model.get_params()['order']
print('ARIMA order:', order, '\n')
# Genereate predictions
prediction = []
for i in range(len(y_test_list)):
model = pmdarima.ARIMA(order=order)
model.fit(y_train_list)
# print('working on', i+1, 'of', len(y_test), '-- ' + str(int(100 * (i + 1) / len(y_test))) + '% complete')
prediction.append(model.predict()[0])
y_train_list.append(y_test_list[i])
predictionte = y_scaler.inverse_transform(np.array(prediction).reshape(-1,1))
y_test_ = y_scaler.inverse_transform(np.array(y_test_list).reshape(-1,1))
# Generate error data
mse = mean_squared_error(yc_test, predictionte)
rmse = mse ** 0.5
mae = mean_absolute_error(y_test_ , predictionte )
return yc,predictionte.flatten().tolist(), mse, rmse, mae
In [ ]:
def get_lstm(data,original_data, train_len, test_len,img_file,ma ,lstm_len=3):
# prepare train and test data
X_value = pd.DataFrame(data.iloc[:, :])
y_value = pd.DataFrame(data.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
# Get data and check shape
X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X)
y_train, y_test, = split_train_test(y)
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
det =20
input_dim = X_train.shape[1]#3
feature_size = X_train.shape[2]#24
output_dim = y_train.shape[1]#1
# Option 1
# Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(256, activation='relu', kernel_initializer='he_normal', input_shape=(input_dim, feature_size)))
# model.add(Dense(units=64,activation='relu'))
# model.add(Dropout(0.5))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(learning_rate = 0.001), loss='mse')
# ## Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# # # option 2
# model = Sequential()
# model.add(Bidirectional(LSTM(units= 128), input_shape=(input_dim, feature_size)))
# model.add(Dense(64))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(lr = 0.001), loss='mean_squared_error', metrics=['accuracy'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM7.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma+' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Option 3
# define custom activation
#
class Double_Tanh(Activation):
def __init__(self, activation, **kwargs):
super(Double_Tanh, self).__init__(activation, **kwargs)
self.__name__ = 'double_tanh'
def double_tanh(x):
return (K.tanh(x) * 2)
get_custom_objects().update({'double_tanh':Double_Tanh(double_tanh)})
# Model Generation
model = Sequential()
#check https://machinelearningmastery.com/use-weight-regularization-lstm-networks-time-series-forecasting/
model.add(LSTM(25, input_shape=(input_dim, feature_size), dropout=0.2, kernel_regularizer=l1_l2(0.00,0.00), bias_regularizer=l1_l2(0.00,0.00)))
model.add(Dense(1))
model.add(Activation(double_tanh))
model.compile(loss='mean_squared_error', optimizer='adam', metrics=['mse', 'mae'])
# Common code
callbacks = [
EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
ModelCheckpoint('LSTM7.h5', verbose=1, save_best_only=True, save_weights_only=True)]
fname1 = img_file+'.png'
tensorflow.keras.utils.plot_model(
model, to_file=fname1, show_shapes=True, show_dtype=False,
show_layer_names=True, expand_nested=False, dpi=96,
layer_range=None, show_layer_activations=False
)
history = model.fit(X_train, y_train, epochs=500, batch_size=int( optimized_period[ma]), verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# plot loss
fname2 = img_file+'-'+ma
plt.title(img_file+'-'+ma+' Loss')
plt.xlabel("Epochs")
plt.ylabel("Loss")
pyplot.plot(history.history['loss'], label='train')
pyplot.plot(history.history['val_loss'], label='validation')
pyplot.legend()
pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# Option 4
# Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(units=lstm_len, return_sequences=True, input_shape=(x_train.shape[1], 1)))
# model.add(LSTM(units=int(lstm_len/2)))
# model.add(Dense(1, activation='sigmoid'))
# model.compile(loss='mean_squared_error', optimizer='adam')
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Generate predictions
predictiontr = model.predict(X_train, verbose=0)
predictiontr = y_scaler.inverse_transform(predictiontr).tolist()
outputtr = []
for i in range(len(predictiontr)):
outputtr.extend(predictiontr[i])
predictiontr = outputtr
# Generate error data
## replace with yc , xtest generated by new multistep method
mse_tr = mean_squared_error(y_train, predictiontr)
rmse_tr = mse_tr ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictiontr))
mae_tr = mean_absolute_error(y_train, pd.Series(predictiontr))
# Original_tr = pd.Series(yc_train)
Original_tr = y_scaler.inverse_transform(y_train).flatten().tolist()
predictionte = model.predict(X_test, verbose=0)
predictionte = (y_scaler.inverse_transform(predictionte)-det).tolist()
outputte = []
for i in range(len(predictionte)):
outputte.extend(predictionte[i])
predictionte = outputte
# Generate error data
mse_te = mean_squared_error(y_test, predictionte)
rmse_te = mse_te ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictionte))
mae_te = mean_absolute_error(y_test, pd.Series(predictionte))
# Original_te = pd.Series(yc_test)
Original_te = y_scaler.inverse_transform(y_test).flatten().tolist()
return Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,Original_te,predictionte, mse_te,rmse_te,mae_te
In [ ]:
if __name__ == '__main__':
start_time = timeit.default_timer()
simulation7 = {}
imgfile = 'Experiment7'
for ma in optimized_period:
print(ma)
print(functions[ma])
print ( int( optimized_period[ma]))
# if ma == 'SMA':
low_vol = df.apply(lambda c: functions[ma](c, timeperiod = int( optimized_period[ma])))
low_vol = low_vol.fillna(0)
low_vol_data = df['close']
high_vol = pd.DataFrame()
df2 = df.copy()
for i in df2.columns:
if i in low_vol.columns:
high_vol[i] = df2[i].subtract(low_vol[i], fill_value=0)
high_vol_data = df['close']
## *****************************************************
# Generate ARIMA and LSTM predictions
print('\nWorking on ' + ma + ' predictions')
try:
print('parameters used : ', train_len, test_len)
low_vol_Original, low_vol_prediction, low_vol_mse, low_vol_rmse,low_vol_mae = get_arima_exog(low_vol,low_vol_data, train_len, test_len)
except:
print('ARIMA error, skipping to next MA type')
continue
Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,high_vol_Original, high_vol_prediction, high_vol_mse, high_vol_rmse,high_vol_mae, = get_lstm(high_vol,high_vol_data, train_len, test_len,imgfile,ma)
final_prediction_tr = df['close'].head(train_len).values + pd.Series(predictiontr) # ignoring first 3 steps
mse_ftr = mean_squared_error(df['close'].head(train_len).values,final_prediction_tr.values)
rmse_ftr = mse_ftr ** 0.5
mape_ftr = mean_absolute_percentage_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
mae_ftr = mean_absolute_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
final_prediction = pd.Series(low_vol_prediction[3:]) + pd.Series(high_vol_prediction)
mse = mean_squared_error(df['close'].tail(test_len).values,final_prediction.values)
rmse = mse ** 0.5
mape = mean_absolute_percentage_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
mae = mean_absolute_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
# Generate prediction accuracy
actual = df['close'].tail(test_len).values
result_1 = []
result_2 = []
for i in range(1, len(final_prediction)):
# Compare prediction to previous close price
if final_prediction[i] > actual[i-1] and actual[i] > actual[i-1]:
result_1.append(1)
elif final_prediction[i] < actual[i-1] and actual[i] < actual[i-1]:
result_1.append(1)
else:
result_1.append(0)
# Compare prediction to previous prediction
if final_prediction[i] > final_prediction[i-1] and actual[i] > actual[i-1]:
result_2.append(1)
elif final_prediction[i] < final_prediction[i-1] and actual[i] < actual[i-1]:
result_2.append(1)
else:
result_2.append(0)
accuracy_1 = np.mean(result_1)
accuracy_2 = np.mean(result_2)
simulation7[ma] = {'low_vol': {'original':list(low_vol_Original), 'prediction': list(low_vol_prediction) , 'mse': low_vol_mse,
'rmse': low_vol_rmse, 'mae' : low_vol_mae},
'high_vol': {'original':list(high_vol_Original),'prediction': list(high_vol_prediction), 'mse': high_vol_mse,
'rmse': high_vol_rmse, 'mae' : high_vol_mae},
'final_tr': {'original':df['close'].head(train_len).tolist(),'prediction': final_prediction_tr.values.tolist(), 'mse': mse_ftr,
'rmse': rmse_ftr, 'mae' : mae_ftr},
'final': {'original': df['close'].tail(test_len).tolist(), 'prediction': final_prediction.values.tolist(), 'mse': mse,
'rmse': rmse, 'mae': mae },
'accuracy': {'prediction vs close': accuracy_1, 'prediction vs prediction': accuracy_2}}
# save simulation data here as checkpoint
with open('simulation7_data.json', 'w') as fp:
json.dump(simulation7, fp)
for ma in simulation7.keys():
print('\n' + ma)
print('Prediction vs Close:\t\t' + str(round(100*simulation7[ma]['accuracy']['prediction vs close'], 2))
+ '% Accuracy')
print('Prediction vs Prediction:\t' + str(round(100*simulation7[ma]['accuracy']['prediction vs prediction'], 2))
+ '% Accuracy')
print('MSE:\t', simulation7[ma]['final']['mse'],
'\nRMSE:\t', simulation7[ma]['final']['rmse'],
'\nMAPE:\t', simulation7[ma]['final']['mae'])#,
# '\nMAPE:\t', simulation[ma]['final']['mape'])
# else:
# break
elapsed = timeit.default_timer() - start_time
print('Runtime: mins:',elapsed/60)
SMA
SMA([input_arrays], [timeperiod=30])
Simple Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
17
Working on SMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-15000.708, Time=8.16 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-13492.284, Time=2.28 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15827.971, Time=8.01 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-13635.197, Time=9.85 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-14132.778, Time=3.78 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15140.312, Time=9.83 sec
ARIMA(1,3,0)(0,0,0)[0] intercept : AIC=-13970.469, Time=7.14 sec
Best model: ARIMA(1,3,0)(0,0,0)[0]
Total fit time: 49.071 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(1, 3, 0) Log Likelihood 7936.985
Date: Sun, 12 Dec 2021 AIC -15827.971
Time: 15:03:36 BIC -15720.081
Sample: 0 HQIC -15786.537
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -4.786e-05 0.001 -0.066 0.947 -0.001 0.001
x2 -4.789e-05 0.001 -0.085 0.932 -0.001 0.001
x3 -4.819e-05 0.000 -0.105 0.917 -0.001 0.001
x4 1.0000 0.001 1557.248 0.000 0.999 1.001
x5 -4.579e-05 0.001 -0.071 0.943 -0.001 0.001
x6 -5.16e-05 0.000 -0.432 0.666 -0.000 0.000
x7 -4.778e-05 0.000 -0.278 0.781 -0.000 0.000
x8 -0.0012 0.000 -7.403 0.000 -0.002 -0.001
x9 -3.454e-06 0.002 -0.002 0.998 -0.003 0.003
x10 -0.0005 0.001 -0.403 0.687 -0.003 0.002
x11 0.0029 0.000 10.904 0.000 0.002 0.003
x12 -0.0003 0.000 -1.815 0.069 -0.001 2.06e-05
x13 -4.809e-05 0.000 -0.157 0.875 -0.001 0.001
x14 -0.0001 0.000 -0.482 0.630 -0.001 0.000
x15 -5.214e-05 0.000 -0.273 0.785 -0.000 0.000
x16 -4.468e-05 0.000 -0.125 0.901 -0.001 0.001
x17 -4.224e-05 0.000 -0.202 0.840 -0.000 0.000
x18 -8.086e-05 0.000 -0.270 0.787 -0.001 0.001
x19 -5.537e-05 0.000 -0.244 0.807 -0.000 0.000
x20 8.423e-05 0.000 0.333 0.739 -0.000 0.001
x21 -4.232e-05 0.000 -0.166 0.868 -0.001 0.000
ar.L1 -0.6666 6.03e-06 -1.11e+05 0.000 -0.667 -0.667
sigma2 4.093e-10 8.97e-11 4.563 0.000 2.33e-10 5.85e-10
===================================================================================
Ljung-Box (L1) (Q): 60.24 Jarque-Bera (JB): 1334882.31
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.11 Skew: -3.81
Prob(H) (two-sided): 0.00 Kurtosis: 202.35
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 5.73e+20. Standard errors may be unstable.
ARIMA order: (1, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.70185, saving model to LSTM7.h5
48/48 - 2s - loss: 0.2036 - mse: 0.2036 - mae: 0.3350 - val_loss: 0.7018 - val_mse: 0.7018 - val_mae: 0.7934 - lr: 0.0010 - 2s/epoch - 49ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.70185 to 0.56641, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0343 - mse: 0.0343 - mae: 0.1490 - val_loss: 0.5664 - val_mse: 0.5664 - val_mae: 0.7092 - lr: 0.0010 - 199ms/epoch - 4ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.56641 to 0.49672, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0187 - mse: 0.0187 - mae: 0.1079 - val_loss: 0.4967 - val_mse: 0.4967 - val_mae: 0.6625 - lr: 0.0010 - 185ms/epoch - 4ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.49672 to 0.46591, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0145 - mse: 0.0145 - mae: 0.0925 - val_loss: 0.4659 - val_mse: 0.4659 - val_mae: 0.6413 - lr: 0.0010 - 219ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.46591 to 0.45875, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0116 - mse: 0.0116 - mae: 0.0842 - val_loss: 0.4588 - val_mse: 0.4588 - val_mae: 0.6361 - lr: 0.0010 - 204ms/epoch - 4ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.45875 to 0.41503, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0116 - mse: 0.0116 - mae: 0.0841 - val_loss: 0.4150 - val_mse: 0.4150 - val_mae: 0.6044 - lr: 0.0010 - 193ms/epoch - 4ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.41503
48/48 - 0s - loss: 0.0093 - mse: 0.0093 - mae: 0.0759 - val_loss: 0.4435 - val_mse: 0.4435 - val_mae: 0.6265 - lr: 0.0010 - 191ms/epoch - 4ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.41503
48/48 - 0s - loss: 0.0117 - mse: 0.0117 - mae: 0.0835 - val_loss: 0.4200 - val_mse: 0.4200 - val_mae: 0.6104 - lr: 0.0010 - 180ms/epoch - 4ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.41503
48/48 - 0s - loss: 0.0085 - mse: 0.0085 - mae: 0.0701 - val_loss: 0.4324 - val_mse: 0.4324 - val_mae: 0.6203 - lr: 0.0010 - 180ms/epoch - 4ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.41503 to 0.39748, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0092 - mse: 0.0092 - mae: 0.0748 - val_loss: 0.3975 - val_mse: 0.3975 - val_mae: 0.5940 - lr: 0.0010 - 189ms/epoch - 4ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.39748
48/48 - 0s - loss: 0.0094 - mse: 0.0094 - mae: 0.0744 - val_loss: 0.4509 - val_mse: 0.4509 - val_mae: 0.6359 - lr: 0.0010 - 178ms/epoch - 4ms/step
Epoch 12/500
Epoch 00012: val_loss improved from 0.39748 to 0.39172, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0101 - mse: 0.0101 - mae: 0.0778 - val_loss: 0.3917 - val_mse: 0.3917 - val_mae: 0.5904 - lr: 0.0010 - 208ms/epoch - 4ms/step
Epoch 13/500
Epoch 00013: val_loss improved from 0.39172 to 0.36768, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0090 - mse: 0.0090 - mae: 0.0729 - val_loss: 0.3677 - val_mse: 0.3677 - val_mae: 0.5712 - lr: 0.0010 - 207ms/epoch - 4ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.36768
48/48 - 0s - loss: 0.0108 - mse: 0.0108 - mae: 0.0778 - val_loss: 0.3685 - val_mse: 0.3685 - val_mae: 0.5724 - lr: 0.0010 - 176ms/epoch - 4ms/step
Epoch 15/500
Epoch 00015: val_loss improved from 0.36768 to 0.35888, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0106 - mse: 0.0106 - mae: 0.0798 - val_loss: 0.3589 - val_mse: 0.3589 - val_mae: 0.5645 - lr: 0.0010 - 189ms/epoch - 4ms/step
Epoch 16/500
Epoch 00016: val_loss improved from 0.35888 to 0.35117, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0105 - mse: 0.0105 - mae: 0.0780 - val_loss: 0.3512 - val_mse: 0.3512 - val_mae: 0.5583 - lr: 0.0010 - 194ms/epoch - 4ms/step
Epoch 17/500
Epoch 00017: val_loss improved from 0.35117 to 0.33856, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0112 - mse: 0.0112 - mae: 0.0819 - val_loss: 0.3386 - val_mse: 0.3386 - val_mae: 0.5473 - lr: 0.0010 - 202ms/epoch - 4ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.33856
48/48 - 0s - loss: 0.0122 - mse: 0.0122 - mae: 0.0852 - val_loss: 0.3443 - val_mse: 0.3443 - val_mae: 0.5520 - lr: 0.0010 - 195ms/epoch - 4ms/step
Epoch 19/500
Epoch 00019: val_loss improved from 0.33856 to 0.29144, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0119 - mse: 0.0119 - mae: 0.0870 - val_loss: 0.2914 - val_mse: 0.2914 - val_mae: 0.5037 - lr: 0.0010 - 187ms/epoch - 4ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.29144
48/48 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0952 - val_loss: 0.3920 - val_mse: 0.3920 - val_mae: 0.5911 - lr: 0.0010 - 182ms/epoch - 4ms/step
Epoch 21/500
Epoch 00021: val_loss improved from 0.29144 to 0.25381, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0138 - mse: 0.0138 - mae: 0.0940 - val_loss: 0.2538 - val_mse: 0.2538 - val_mae: 0.4667 - lr: 0.0010 - 192ms/epoch - 4ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.25381
48/48 - 0s - loss: 0.0150 - mse: 0.0150 - mae: 0.0965 - val_loss: 0.3521 - val_mse: 0.3521 - val_mae: 0.5574 - lr: 0.0010 - 193ms/epoch - 4ms/step
Epoch 23/500
Epoch 00023: val_loss improved from 0.25381 to 0.21639, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0136 - mse: 0.0136 - mae: 0.0928 - val_loss: 0.2164 - val_mse: 0.2164 - val_mae: 0.4260 - lr: 0.0010 - 218ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.21639
48/48 - 0s - loss: 0.0137 - mse: 0.0137 - mae: 0.0955 - val_loss: 0.3752 - val_mse: 0.3752 - val_mae: 0.5750 - lr: 0.0010 - 202ms/epoch - 4ms/step
Epoch 25/500
Epoch 00025: val_loss improved from 0.21639 to 0.17944, saving model to LSTM7.h5
48/48 - 0s - loss: 0.0119 - mse: 0.0119 - mae: 0.0876 - val_loss: 0.1794 - val_mse: 0.1794 - val_mae: 0.3819 - lr: 0.0010 - 193ms/epoch - 4ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0105 - mse: 0.0105 - mae: 0.0824 - val_loss: 0.3447 - val_mse: 0.3447 - val_mae: 0.5489 - lr: 0.0010 - 194ms/epoch - 4ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0091 - mse: 0.0091 - mae: 0.0763 - val_loss: 0.1934 - val_mse: 0.1934 - val_mae: 0.3992 - lr: 0.0010 - 187ms/epoch - 4ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0090 - mse: 0.0090 - mae: 0.0761 - val_loss: 0.3444 - val_mse: 0.3444 - val_mae: 0.5488 - lr: 0.0010 - 198ms/epoch - 4ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0087 - mse: 0.0087 - mae: 0.0738 - val_loss: 0.2230 - val_mse: 0.2230 - val_mae: 0.4327 - lr: 0.0010 - 185ms/epoch - 4ms/step
Epoch 30/500
Epoch 00030: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00030: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0087 - mse: 0.0087 - mae: 0.0735 - val_loss: 0.3052 - val_mse: 0.3052 - val_mae: 0.5142 - lr: 0.0010 - 183ms/epoch - 4ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0137 - mse: 0.0137 - mae: 0.0957 - val_loss: 0.2431 - val_mse: 0.2431 - val_mae: 0.4553 - lr: 1.0000e-04 - 182ms/epoch - 4ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0563 - val_loss: 0.2285 - val_mse: 0.2285 - val_mae: 0.4405 - lr: 1.0000e-04 - 178ms/epoch - 4ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0569 - val_loss: 0.2234 - val_mse: 0.2234 - val_mae: 0.4351 - lr: 1.0000e-04 - 188ms/epoch - 4ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0556 - val_loss: 0.2208 - val_mse: 0.2208 - val_mae: 0.4323 - lr: 1.0000e-04 - 187ms/epoch - 4ms/step
Epoch 35/500
Epoch 00035: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00035: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0566 - val_loss: 0.2211 - val_mse: 0.2211 - val_mae: 0.4326 - lr: 1.0000e-04 - 183ms/epoch - 4ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0527 - val_loss: 0.2209 - val_mse: 0.2209 - val_mae: 0.4324 - lr: 1.0000e-05 - 191ms/epoch - 4ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0510 - val_loss: 0.2209 - val_mse: 0.2209 - val_mae: 0.4325 - lr: 1.0000e-05 - 177ms/epoch - 4ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0536 - val_loss: 0.2207 - val_mse: 0.2207 - val_mae: 0.4322 - lr: 1.0000e-05 - 193ms/epoch - 4ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0522 - val_loss: 0.2206 - val_mse: 0.2206 - val_mae: 0.4321 - lr: 1.0000e-05 - 180ms/epoch - 4ms/step
Epoch 40/500
Epoch 00040: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00040: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0522 - val_loss: 0.2209 - val_mse: 0.2209 - val_mae: 0.4325 - lr: 1.0000e-05 - 186ms/epoch - 4ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0543 - val_loss: 0.2209 - val_mse: 0.2209 - val_mae: 0.4325 - lr: 1.0000e-05 - 180ms/epoch - 4ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0536 - val_loss: 0.2205 - val_mse: 0.2205 - val_mae: 0.4320 - lr: 1.0000e-05 - 185ms/epoch - 4ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0520 - val_loss: 0.2204 - val_mse: 0.2204 - val_mae: 0.4320 - lr: 1.0000e-05 - 189ms/epoch - 4ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0520 - val_loss: 0.2209 - val_mse: 0.2209 - val_mae: 0.4325 - lr: 1.0000e-05 - 184ms/epoch - 4ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0535 - val_loss: 0.2209 - val_mse: 0.2209 - val_mae: 0.4324 - lr: 1.0000e-05 - 204ms/epoch - 4ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0547 - val_loss: 0.2206 - val_mse: 0.2206 - val_mae: 0.4322 - lr: 1.0000e-05 - 186ms/epoch - 4ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0522 - val_loss: 0.2206 - val_mse: 0.2206 - val_mae: 0.4321 - lr: 1.0000e-05 - 189ms/epoch - 4ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0522 - val_loss: 0.2208 - val_mse: 0.2208 - val_mae: 0.4323 - lr: 1.0000e-05 - 182ms/epoch - 4ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0517 - val_loss: 0.2212 - val_mse: 0.2212 - val_mae: 0.4327 - lr: 1.0000e-05 - 188ms/epoch - 4ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0528 - val_loss: 0.2217 - val_mse: 0.2217 - val_mae: 0.4333 - lr: 1.0000e-05 - 175ms/epoch - 4ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0512 - val_loss: 0.2219 - val_mse: 0.2219 - val_mae: 0.4335 - lr: 1.0000e-05 - 179ms/epoch - 4ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0512 - val_loss: 0.2216 - val_mse: 0.2216 - val_mae: 0.4332 - lr: 1.0000e-05 - 182ms/epoch - 4ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0539 - val_loss: 0.2210 - val_mse: 0.2210 - val_mae: 0.4325 - lr: 1.0000e-05 - 179ms/epoch - 4ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0517 - val_loss: 0.2211 - val_mse: 0.2211 - val_mae: 0.4326 - lr: 1.0000e-05 - 191ms/epoch - 4ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0524 - val_loss: 0.2217 - val_mse: 0.2217 - val_mae: 0.4332 - lr: 1.0000e-05 - 179ms/epoch - 4ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0515 - val_loss: 0.2219 - val_mse: 0.2219 - val_mae: 0.4335 - lr: 1.0000e-05 - 175ms/epoch - 4ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0047 - mse: 0.0047 - mae: 0.0540 - val_loss: 0.2216 - val_mse: 0.2216 - val_mae: 0.4331 - lr: 1.0000e-05 - 197ms/epoch - 4ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0512 - val_loss: 0.2218 - val_mse: 0.2218 - val_mae: 0.4333 - lr: 1.0000e-05 - 190ms/epoch - 4ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0047 - mse: 0.0047 - mae: 0.0535 - val_loss: 0.2208 - val_mse: 0.2208 - val_mae: 0.4323 - lr: 1.0000e-05 - 189ms/epoch - 4ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0519 - val_loss: 0.2203 - val_mse: 0.2203 - val_mae: 0.4318 - lr: 1.0000e-05 - 186ms/epoch - 4ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0515 - val_loss: 0.2200 - val_mse: 0.2200 - val_mae: 0.4314 - lr: 1.0000e-05 - 187ms/epoch - 4ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0528 - val_loss: 0.2196 - val_mse: 0.2196 - val_mae: 0.4310 - lr: 1.0000e-05 - 181ms/epoch - 4ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0524 - val_loss: 0.2199 - val_mse: 0.2199 - val_mae: 0.4314 - lr: 1.0000e-05 - 184ms/epoch - 4ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0527 - val_loss: 0.2198 - val_mse: 0.2198 - val_mae: 0.4313 - lr: 1.0000e-05 - 203ms/epoch - 4ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0530 - val_loss: 0.2209 - val_mse: 0.2209 - val_mae: 0.4324 - lr: 1.0000e-05 - 190ms/epoch - 4ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0513 - val_loss: 0.2213 - val_mse: 0.2213 - val_mae: 0.4328 - lr: 1.0000e-05 - 186ms/epoch - 4ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0524 - val_loss: 0.2211 - val_mse: 0.2211 - val_mae: 0.4326 - lr: 1.0000e-05 - 186ms/epoch - 4ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0519 - val_loss: 0.2217 - val_mse: 0.2217 - val_mae: 0.4332 - lr: 1.0000e-05 - 190ms/epoch - 4ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0526 - val_loss: 0.2219 - val_mse: 0.2219 - val_mae: 0.4335 - lr: 1.0000e-05 - 194ms/epoch - 4ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0535 - val_loss: 0.2221 - val_mse: 0.2221 - val_mae: 0.4336 - lr: 1.0000e-05 - 200ms/epoch - 4ms/step
Epoch 71/500
Epoch 00071: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0532 - val_loss: 0.2232 - val_mse: 0.2232 - val_mae: 0.4348 - lr: 1.0000e-05 - 184ms/epoch - 4ms/step
Epoch 72/500
Epoch 00072: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0507 - val_loss: 0.2233 - val_mse: 0.2233 - val_mae: 0.4349 - lr: 1.0000e-05 - 179ms/epoch - 4ms/step
Epoch 73/500
Epoch 00073: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0509 - val_loss: 0.2219 - val_mse: 0.2219 - val_mae: 0.4334 - lr: 1.0000e-05 - 194ms/epoch - 4ms/step
Epoch 74/500
Epoch 00074: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0511 - val_loss: 0.2232 - val_mse: 0.2232 - val_mae: 0.4348 - lr: 1.0000e-05 - 186ms/epoch - 4ms/step
Epoch 75/500
Epoch 00075: val_loss did not improve from 0.17944
48/48 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0516 - val_loss: 0.2239 - val_mse: 0.2239 - val_mae: 0.4356 - lr: 1.0000e-05 - 201ms/epoch - 4ms/step
Epoch 00075: early stopping
SMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 37.148636819682245
RMSE: 6.094968155756209
MAPE: 5.090179331518223
EMA
EMA([input_arrays], [timeperiod=30])
Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
51
Working on EMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.775, Time=2.31 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=3.91 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16801.081, Time=8.99 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.31 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-14532.068, Time=7.29 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15610.472, Time=12.83 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16103.302, Time=13.06 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17030.021, Time=4.15 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.614, Time=3.48 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.441, Time=6.61 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=16.57 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17003.984, Time=3.00 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-16998.666, Time=3.68 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 91.197 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.720
Date: Sun, 12 Dec 2021 AIC -17087.441
Time: 15:06:04 BIC -16965.479
Sample: 0 HQIC -17040.602
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.316e-10 9.87e-21 -2.35e+10 0.000 -2.32e-10 -2.32e-10
x2 -2.308e-10 9.85e-21 -2.34e+10 0.000 -2.31e-10 -2.31e-10
x3 -2.324e-10 9.88e-21 -2.35e+10 0.000 -2.32e-10 -2.32e-10
x4 1.0000 9.87e-21 1.01e+20 0.000 1.000 1.000
x5 -2.106e-10 9.41e-21 -2.24e+10 0.000 -2.11e-10 -2.11e-10
x6 -7.996e-10 1.74e-20 -4.59e+10 0.000 -8e-10 -8e-10
x7 -2.295e-10 9.82e-21 -2.34e+10 0.000 -2.29e-10 -2.29e-10
x8 -2.244e-10 9.71e-21 -2.31e+10 0.000 -2.24e-10 -2.24e-10
x9 -1.166e-11 1.98e-21 -5.9e+09 0.000 -1.17e-11 -1.17e-11
x10 -4.453e-11 4.22e-21 -1.06e+10 0.000 -4.45e-11 -4.45e-11
x11 -2.219e-10 9.65e-21 -2.3e+10 0.000 -2.22e-10 -2.22e-10
x12 -2.264e-10 9.76e-21 -2.32e+10 0.000 -2.26e-10 -2.26e-10
x13 -2.315e-10 9.87e-21 -2.35e+10 0.000 -2.31e-10 -2.31e-10
x14 -1.766e-09 2.73e-20 -6.48e+10 0.000 -1.77e-09 -1.77e-09
x15 -2.167e-10 9.37e-21 -2.31e+10 0.000 -2.17e-10 -2.17e-10
x16 -5.232e-10 1.49e-20 -3.52e+10 0.000 -5.23e-10 -5.23e-10
x17 -2.147e-10 9.48e-21 -2.27e+10 0.000 -2.15e-10 -2.15e-10
x18 -3.791e-11 3.96e-21 -9.56e+09 0.000 -3.79e-11 -3.79e-11
x19 -2.597e-10 1.05e-20 -2.48e+10 0.000 -2.6e-10 -2.6e-10
x20 -2.417e-10 1e-20 -2.41e+10 0.000 -2.42e-10 -2.42e-10
x21 -4.823e-10 1.4e-20 -3.44e+10 0.000 -4.82e-10 -4.82e-10
ar.L1 -0.4923 1.46e-22 -3.38e+21 0.000 -0.492 -0.492
ar.L2 -0.1923 8.47e-23 -2.27e+21 0.000 -0.192 -0.192
ar.L3 -0.0462 4.02e-23 -1.15e+21 0.000 -0.046 -0.046
ma.L1 -0.7077 3.31e-22 -2.14e+21 0.000 -0.708 -0.708
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 54.09 Jarque-Bera (JB): 4207353.17
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.48
Prob(H) (two-sided): 0.00 Kurtosis: 357.00
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 3.15e+40. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.05006, saving model to LSTM7.h5
16/16 - 2s - loss: 0.2215 - mse: 0.2215 - mae: 0.3350 - val_loss: 0.0501 - val_mse: 0.0501 - val_mae: 0.1733 - lr: 0.0010 - 2s/epoch - 134ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.05006
16/16 - 0s - loss: 0.0689 - mse: 0.0689 - mae: 0.2211 - val_loss: 0.0511 - val_mse: 0.0511 - val_mae: 0.1710 - lr: 0.0010 - 78ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.05006
16/16 - 0s - loss: 0.0288 - mse: 0.0288 - mae: 0.1337 - val_loss: 0.0510 - val_mse: 0.0510 - val_mae: 0.1698 - lr: 0.0010 - 87ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.05006 to 0.04261, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0260 - mse: 0.0260 - mae: 0.1281 - val_loss: 0.0426 - val_mse: 0.0426 - val_mae: 0.1669 - lr: 0.0010 - 96ms/epoch - 6ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.04261 to 0.04178, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0194 - mse: 0.0194 - mae: 0.1136 - val_loss: 0.0418 - val_mse: 0.0418 - val_mae: 0.1637 - lr: 0.0010 - 96ms/epoch - 6ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.04178 to 0.04063, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0179 - mse: 0.0179 - mae: 0.1068 - val_loss: 0.0406 - val_mse: 0.0406 - val_mae: 0.1600 - lr: 0.0010 - 88ms/epoch - 6ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.04063 to 0.04050, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0963 - val_loss: 0.0405 - val_mse: 0.0405 - val_mae: 0.1560 - lr: 0.0010 - 85ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.04050 to 0.03978, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0142 - mse: 0.0142 - mae: 0.0945 - val_loss: 0.0398 - val_mse: 0.0398 - val_mae: 0.1534 - lr: 0.0010 - 95ms/epoch - 6ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.03978 to 0.03869, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0134 - mse: 0.0134 - mae: 0.0928 - val_loss: 0.0387 - val_mse: 0.0387 - val_mae: 0.1508 - lr: 0.0010 - 84ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.03869 to 0.03820, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0116 - mse: 0.0116 - mae: 0.0863 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1486 - lr: 0.0010 - 86ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: val_loss improved from 0.03820 to 0.03602, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0112 - mse: 0.0112 - mae: 0.0833 - val_loss: 0.0360 - val_mse: 0.0360 - val_mae: 0.1464 - lr: 0.0010 - 89ms/epoch - 6ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.03602
16/16 - 0s - loss: 0.0106 - mse: 0.0106 - mae: 0.0824 - val_loss: 0.0366 - val_mse: 0.0366 - val_mae: 0.1438 - lr: 0.0010 - 72ms/epoch - 4ms/step
Epoch 13/500
Epoch 00013: val_loss improved from 0.03602 to 0.03489, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0110 - mse: 0.0110 - mae: 0.0830 - val_loss: 0.0349 - val_mse: 0.0349 - val_mae: 0.1405 - lr: 0.0010 - 92ms/epoch - 6ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.03489
16/16 - 0s - loss: 0.0096 - mse: 0.0096 - mae: 0.0762 - val_loss: 0.0361 - val_mse: 0.0361 - val_mae: 0.1390 - lr: 0.0010 - 86ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss improved from 0.03489 to 0.03484, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0093 - mse: 0.0093 - mae: 0.0761 - val_loss: 0.0348 - val_mse: 0.0348 - val_mae: 0.1364 - lr: 0.0010 - 101ms/epoch - 6ms/step
Epoch 16/500
Epoch 00016: val_loss improved from 0.03484 to 0.03461, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0088 - mse: 0.0088 - mae: 0.0755 - val_loss: 0.0346 - val_mse: 0.0346 - val_mae: 0.1346 - lr: 0.0010 - 88ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.03461
16/16 - 0s - loss: 0.0085 - mse: 0.0085 - mae: 0.0727 - val_loss: 0.0363 - val_mse: 0.0363 - val_mae: 0.1351 - lr: 0.0010 - 80ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss improved from 0.03461 to 0.03449, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0710 - val_loss: 0.0345 - val_mse: 0.0345 - val_mae: 0.1322 - lr: 0.0010 - 109ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.03449
16/16 - 0s - loss: 0.0074 - mse: 0.0074 - mae: 0.0681 - val_loss: 0.0351 - val_mse: 0.0351 - val_mae: 0.1324 - lr: 0.0010 - 76ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss improved from 0.03449 to 0.03434, saving model to LSTM7.h5
16/16 - 0s - loss: 0.0079 - mse: 0.0079 - mae: 0.0711 - val_loss: 0.0343 - val_mse: 0.0343 - val_mae: 0.1310 - lr: 0.0010 - 87ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0649 - val_loss: 0.0361 - val_mse: 0.0361 - val_mae: 0.1337 - lr: 0.0010 - 77ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0074 - mse: 0.0074 - mae: 0.0685 - val_loss: 0.0365 - val_mse: 0.0365 - val_mae: 0.1346 - lr: 0.0010 - 74ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0647 - val_loss: 0.0372 - val_mse: 0.0372 - val_mae: 0.1362 - lr: 0.0010 - 71ms/epoch - 4ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0635 - val_loss: 0.0374 - val_mse: 0.0374 - val_mae: 0.1373 - lr: 0.0010 - 71ms/epoch - 4ms/step
Epoch 25/500
Epoch 00025: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00025: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0636 - val_loss: 0.0378 - val_mse: 0.0378 - val_mae: 0.1387 - lr: 0.0010 - 84ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0588 - val_loss: 0.0375 - val_mse: 0.0375 - val_mae: 0.1379 - lr: 1.0000e-04 - 83ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0577 - val_loss: 0.0370 - val_mse: 0.0370 - val_mae: 0.1371 - lr: 1.0000e-04 - 78ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0587 - val_loss: 0.0369 - val_mse: 0.0369 - val_mae: 0.1368 - lr: 1.0000e-04 - 80ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0599 - val_loss: 0.0370 - val_mse: 0.0370 - val_mae: 0.1371 - lr: 1.0000e-04 - 78ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00030: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0617 - val_loss: 0.0371 - val_mse: 0.0371 - val_mae: 0.1373 - lr: 1.0000e-04 - 72ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0612 - val_loss: 0.0371 - val_mse: 0.0371 - val_mae: 0.1373 - lr: 1.0000e-05 - 74ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0595 - val_loss: 0.0371 - val_mse: 0.0371 - val_mae: 0.1373 - lr: 1.0000e-05 - 75ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0611 - val_loss: 0.0371 - val_mse: 0.0371 - val_mae: 0.1374 - lr: 1.0000e-05 - 77ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0573 - val_loss: 0.0371 - val_mse: 0.0371 - val_mae: 0.1374 - lr: 1.0000e-05 - 75ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00035: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0567 - val_loss: 0.0372 - val_mse: 0.0372 - val_mae: 0.1375 - lr: 1.0000e-05 - 76ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0604 - val_loss: 0.0372 - val_mse: 0.0372 - val_mae: 0.1376 - lr: 1.0000e-05 - 74ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0578 - val_loss: 0.0373 - val_mse: 0.0373 - val_mae: 0.1377 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0577 - val_loss: 0.0373 - val_mse: 0.0373 - val_mae: 0.1378 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0575 - val_loss: 0.0374 - val_mse: 0.0374 - val_mae: 0.1380 - lr: 1.0000e-05 - 83ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0586 - val_loss: 0.0374 - val_mse: 0.0374 - val_mae: 0.1381 - lr: 1.0000e-05 - 76ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0595 - val_loss: 0.0374 - val_mse: 0.0374 - val_mae: 0.1381 - lr: 1.0000e-05 - 79ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0579 - val_loss: 0.0375 - val_mse: 0.0375 - val_mae: 0.1382 - lr: 1.0000e-05 - 76ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0616 - val_loss: 0.0375 - val_mse: 0.0375 - val_mae: 0.1382 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0569 - val_loss: 0.0375 - val_mse: 0.0375 - val_mae: 0.1383 - lr: 1.0000e-05 - 74ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0591 - val_loss: 0.0375 - val_mse: 0.0375 - val_mae: 0.1384 - lr: 1.0000e-05 - 78ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0589 - val_loss: 0.0376 - val_mse: 0.0376 - val_mae: 0.1384 - lr: 1.0000e-05 - 78ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0549 - val_loss: 0.0376 - val_mse: 0.0376 - val_mae: 0.1384 - lr: 1.0000e-05 - 77ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0573 - val_loss: 0.0375 - val_mse: 0.0375 - val_mae: 0.1384 - lr: 1.0000e-05 - 79ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0587 - val_loss: 0.0376 - val_mse: 0.0376 - val_mae: 0.1384 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0585 - val_loss: 0.0376 - val_mse: 0.0376 - val_mae: 0.1385 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0589 - val_loss: 0.0376 - val_mse: 0.0376 - val_mae: 0.1385 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0574 - val_loss: 0.0376 - val_mse: 0.0376 - val_mae: 0.1386 - lr: 1.0000e-05 - 79ms/epoch - 5ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0604 - val_loss: 0.0376 - val_mse: 0.0376 - val_mae: 0.1386 - lr: 1.0000e-05 - 80ms/epoch - 5ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0590 - val_loss: 0.0377 - val_mse: 0.0377 - val_mae: 0.1387 - lr: 1.0000e-05 - 71ms/epoch - 4ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0581 - val_loss: 0.0377 - val_mse: 0.0377 - val_mae: 0.1388 - lr: 1.0000e-05 - 76ms/epoch - 5ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0577 - val_loss: 0.0377 - val_mse: 0.0377 - val_mae: 0.1388 - lr: 1.0000e-05 - 77ms/epoch - 5ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0621 - val_loss: 0.0377 - val_mse: 0.0377 - val_mae: 0.1388 - lr: 1.0000e-05 - 83ms/epoch - 5ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0586 - val_loss: 0.0377 - val_mse: 0.0377 - val_mae: 0.1388 - lr: 1.0000e-05 - 75ms/epoch - 5ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0594 - val_loss: 0.0377 - val_mse: 0.0377 - val_mae: 0.1388 - lr: 1.0000e-05 - 77ms/epoch - 5ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0569 - val_loss: 0.0377 - val_mse: 0.0377 - val_mae: 0.1389 - lr: 1.0000e-05 - 83ms/epoch - 5ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0557 - val_loss: 0.0378 - val_mse: 0.0378 - val_mae: 0.1390 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0598 - val_loss: 0.0378 - val_mse: 0.0378 - val_mae: 0.1390 - lr: 1.0000e-05 - 81ms/epoch - 5ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0572 - val_loss: 0.0378 - val_mse: 0.0378 - val_mae: 0.1390 - lr: 1.0000e-05 - 82ms/epoch - 5ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0579 - val_loss: 0.0378 - val_mse: 0.0378 - val_mae: 0.1392 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0560 - val_loss: 0.0379 - val_mse: 0.0379 - val_mae: 0.1392 - lr: 1.0000e-05 - 79ms/epoch - 5ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0583 - val_loss: 0.0379 - val_mse: 0.0379 - val_mae: 0.1394 - lr: 1.0000e-05 - 73ms/epoch - 5ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0587 - val_loss: 0.0379 - val_mse: 0.0379 - val_mae: 0.1394 - lr: 1.0000e-05 - 78ms/epoch - 5ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0588 - val_loss: 0.0379 - val_mse: 0.0379 - val_mae: 0.1394 - lr: 1.0000e-05 - 79ms/epoch - 5ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0050 - mse: 0.0050 - mae: 0.0557 - val_loss: 0.0380 - val_mse: 0.0380 - val_mae: 0.1395 - lr: 1.0000e-05 - 79ms/epoch - 5ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.03434
16/16 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0576 - val_loss: 0.0379 - val_mse: 0.0379 - val_mae: 0.1395 - lr: 1.0000e-05 - 74ms/epoch - 5ms/step
Epoch 00070: early stopping
SMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 37.148636819682245
RMSE: 6.094968155756209
MAPE: 5.090179331518223
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 47.39% Accuracy
MSE: 31.582654654902484
RMSE: 5.619844718041815
MAPE: 4.507182634072088
WMA
WMA([input_arrays], [timeperiod=30])
Weighted Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
49
Working on WMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-14480.432, Time=9.44 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-15747.905, Time=6.44 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15116.389, Time=7.16 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-13532.115, Time=7.81 sec
ARIMA(0,3,0)(0,0,0)[0] intercept : AIC=-13619.624, Time=5.40 sec
Best model: ARIMA(0,3,0)(0,0,0)[0]
Total fit time: 36.273 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(0, 3, 0) Log Likelihood 7895.952
Date: Sun, 12 Dec 2021 AIC -15747.905
Time: 15:14:07 BIC -15644.706
Sample: 0 HQIC -15708.272
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 3.384e-05 1.9e-05 1.778 0.075 -3.47e-06 7.12e-05
x2 3.379e-05 1.84e-05 1.832 0.067 -2.35e-06 6.99e-05
x3 3.388e-05 4.34e-05 0.781 0.435 -5.12e-05 0.000
x4 1.0000 4.12e-06 2.43e+05 0.000 1.000 1.000
x5 3.227e-05 3.52e-06 9.163 0.000 2.54e-05 3.92e-05
x6 5.559e-05 6.75e-05 0.823 0.410 -7.67e-05 0.000
x7 3.369e-05 2.38e-05 1.415 0.157 -1.3e-05 8.03e-05
x8 0.0023 2.6e-05 86.661 0.000 0.002 0.002
x9 -8.72e-06 7.51e-07 -11.610 0.000 -1.02e-05 -7.25e-06
x10 -0.0023 3.33e-05 -67.770 0.000 -0.002 -0.002
x11 0.0093 2.8e-05 333.459 0.000 0.009 0.009
x12 -0.0118 2.37e-05 -498.171 0.000 -0.012 -0.012
x13 3.382e-05 1.49e-05 2.273 0.023 4.66e-06 6.3e-05
x14 9.271e-05 6.21e-05 1.493 0.135 -2.9e-05 0.000
x15 3.096e-05 1.92e-05 1.614 0.106 -6.63e-06 6.86e-05
x16 5.52e-05 7.17e-05 0.770 0.441 -8.53e-05 0.000
x17 3.38e-05 3.2e-05 1.056 0.291 -2.89e-05 9.65e-05
x18 -6.715e-06 8.34e-05 -0.081 0.936 -0.000 0.000
x19 3.428e-05 2.07e-05 1.654 0.098 -6.34e-06 7.49e-05
x20 -8.089e-06 9.55e-05 -0.085 0.933 -0.000 0.000
x21 4.255e-05 0.000 0.094 0.925 -0.001 0.001
sigma2 2.581e-10 7.87e-11 3.280 0.001 1.04e-10 4.12e-10
===================================================================================
Ljung-Box (L1) (Q): 362.92 Jarque-Bera (JB): 5047564.68
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.01 Skew: -11.23
Prob(H) (two-sided): 0.00 Kurtosis: 390.28
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.75e+20. Standard errors may be unstable.
ARIMA order: (0, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.39083, saving model to LSTM7.h5
17/17 - 3s - loss: 0.8312 - mse: 0.8312 - mae: 0.7302 - val_loss: 0.3908 - val_mse: 0.3908 - val_mae: 0.5818 - lr: 0.0010 - 3s/epoch - 155ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.39083 to 0.24821, saving model to LSTM7.h5
17/17 - 0s - loss: 0.0523 - mse: 0.0523 - mae: 0.1902 - val_loss: 0.2482 - val_mse: 0.2482 - val_mae: 0.4551 - lr: 0.0010 - 91ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0369 - mse: 0.0369 - mae: 0.1630 - val_loss: 0.3136 - val_mse: 0.3136 - val_mae: 0.5155 - lr: 0.0010 - 86ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0196 - mse: 0.0196 - mae: 0.1106 - val_loss: 0.3367 - val_mse: 0.3367 - val_mae: 0.5347 - lr: 0.0010 - 87ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0230 - mse: 0.0230 - mae: 0.1211 - val_loss: 0.2966 - val_mse: 0.2966 - val_mae: 0.4981 - lr: 0.0010 - 86ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0189 - mse: 0.0189 - mae: 0.1090 - val_loss: 0.2815 - val_mse: 0.2815 - val_mae: 0.4831 - lr: 0.0010 - 90ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00007: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0188 - mse: 0.0188 - mae: 0.1074 - val_loss: 0.3017 - val_mse: 0.3017 - val_mae: 0.5011 - lr: 0.0010 - 86ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0157 - mse: 0.0157 - mae: 0.0999 - val_loss: 0.3026 - val_mse: 0.3026 - val_mae: 0.5019 - lr: 1.0000e-04 - 82ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0178 - mse: 0.0178 - mae: 0.1052 - val_loss: 0.3013 - val_mse: 0.3013 - val_mae: 0.5006 - lr: 1.0000e-04 - 84ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0165 - mse: 0.0165 - mae: 0.1009 - val_loss: 0.3016 - val_mse: 0.3016 - val_mae: 0.5008 - lr: 1.0000e-04 - 80ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0165 - mse: 0.0165 - mae: 0.1024 - val_loss: 0.3005 - val_mse: 0.3005 - val_mae: 0.4998 - lr: 1.0000e-04 - 80ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00012: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0168 - mse: 0.0168 - mae: 0.1029 - val_loss: 0.3009 - val_mse: 0.3009 - val_mae: 0.5000 - lr: 1.0000e-04 - 83ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0167 - mse: 0.0167 - mae: 0.0996 - val_loss: 0.3009 - val_mse: 0.3009 - val_mae: 0.5001 - lr: 1.0000e-05 - 79ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0164 - mse: 0.0164 - mae: 0.1001 - val_loss: 0.3009 - val_mse: 0.3009 - val_mae: 0.5000 - lr: 1.0000e-05 - 91ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0150 - mse: 0.0150 - mae: 0.0963 - val_loss: 0.3007 - val_mse: 0.3007 - val_mae: 0.4999 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0170 - mse: 0.0170 - mae: 0.1022 - val_loss: 0.3007 - val_mse: 0.3007 - val_mae: 0.4999 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00017: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.1002 - val_loss: 0.3010 - val_mse: 0.3010 - val_mae: 0.5002 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0165 - mse: 0.0165 - mae: 0.1006 - val_loss: 0.3010 - val_mse: 0.3010 - val_mae: 0.5001 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.1004 - val_loss: 0.3009 - val_mse: 0.3009 - val_mae: 0.5000 - lr: 1.0000e-05 - 77ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0154 - mse: 0.0154 - mae: 0.0980 - val_loss: 0.3008 - val_mse: 0.3008 - val_mae: 0.4999 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0162 - mse: 0.0162 - mae: 0.1005 - val_loss: 0.3008 - val_mse: 0.3008 - val_mae: 0.4999 - lr: 1.0000e-05 - 88ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0136 - mse: 0.0136 - mae: 0.0947 - val_loss: 0.3008 - val_mse: 0.3008 - val_mae: 0.4999 - lr: 1.0000e-05 - 80ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0166 - mse: 0.0166 - mae: 0.1007 - val_loss: 0.3006 - val_mse: 0.3006 - val_mae: 0.4997 - lr: 1.0000e-05 - 78ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0154 - mse: 0.0154 - mae: 0.0985 - val_loss: 0.3007 - val_mse: 0.3007 - val_mae: 0.4998 - lr: 1.0000e-05 - 81ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0150 - mse: 0.0150 - mae: 0.0966 - val_loss: 0.3008 - val_mse: 0.3008 - val_mae: 0.4998 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0149 - mse: 0.0149 - mae: 0.0965 - val_loss: 0.3009 - val_mse: 0.3009 - val_mae: 0.5000 - lr: 1.0000e-05 - 91ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0169 - mse: 0.0169 - mae: 0.1035 - val_loss: 0.3006 - val_mse: 0.3006 - val_mae: 0.4997 - lr: 1.0000e-05 - 101ms/epoch - 6ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0157 - mse: 0.0157 - mae: 0.0997 - val_loss: 0.3005 - val_mse: 0.3005 - val_mae: 0.4995 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0153 - mse: 0.0153 - mae: 0.0973 - val_loss: 0.3007 - val_mse: 0.3007 - val_mae: 0.4997 - lr: 1.0000e-05 - 83ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0183 - mse: 0.0183 - mae: 0.1053 - val_loss: 0.3004 - val_mse: 0.3004 - val_mae: 0.4995 - lr: 1.0000e-05 - 83ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0154 - mse: 0.0154 - mae: 0.0975 - val_loss: 0.3003 - val_mse: 0.3003 - val_mae: 0.4994 - lr: 1.0000e-05 - 81ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.0993 - val_loss: 0.3002 - val_mse: 0.3002 - val_mae: 0.4992 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0162 - mse: 0.0162 - mae: 0.1020 - val_loss: 0.2999 - val_mse: 0.2999 - val_mae: 0.4990 - lr: 1.0000e-05 - 78ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.1016 - val_loss: 0.3000 - val_mse: 0.3000 - val_mae: 0.4990 - lr: 1.0000e-05 - 81ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0154 - mse: 0.0154 - mae: 0.0968 - val_loss: 0.2997 - val_mse: 0.2997 - val_mae: 0.4987 - lr: 1.0000e-05 - 82ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.0991 - val_loss: 0.2996 - val_mse: 0.2996 - val_mae: 0.4986 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0153 - mse: 0.0153 - mae: 0.0971 - val_loss: 0.2996 - val_mse: 0.2996 - val_mae: 0.4986 - lr: 1.0000e-05 - 90ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0175 - mse: 0.0175 - mae: 0.1026 - val_loss: 0.2994 - val_mse: 0.2994 - val_mae: 0.4984 - lr: 1.0000e-05 - 90ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0152 - mse: 0.0152 - mae: 0.0980 - val_loss: 0.2994 - val_mse: 0.2994 - val_mae: 0.4984 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0171 - mse: 0.0171 - mae: 0.1039 - val_loss: 0.2994 - val_mse: 0.2994 - val_mae: 0.4984 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0144 - mse: 0.0144 - mae: 0.0951 - val_loss: 0.2999 - val_mse: 0.2999 - val_mae: 0.4988 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0166 - mse: 0.0166 - mae: 0.1000 - val_loss: 0.3002 - val_mse: 0.3002 - val_mae: 0.4991 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.0997 - val_loss: 0.3001 - val_mse: 0.3001 - val_mae: 0.4990 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0175 - mse: 0.0175 - mae: 0.1047 - val_loss: 0.3001 - val_mse: 0.3001 - val_mae: 0.4990 - lr: 1.0000e-05 - 82ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.0972 - val_loss: 0.3002 - val_mse: 0.3002 - val_mae: 0.4991 - lr: 1.0000e-05 - 83ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0146 - mse: 0.0146 - mae: 0.0943 - val_loss: 0.3001 - val_mse: 0.3001 - val_mae: 0.4990 - lr: 1.0000e-05 - 79ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0157 - mse: 0.0157 - mae: 0.0962 - val_loss: 0.2999 - val_mse: 0.2999 - val_mae: 0.4988 - lr: 1.0000e-05 - 79ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0157 - mse: 0.0157 - mae: 0.1003 - val_loss: 0.3000 - val_mse: 0.3000 - val_mae: 0.4988 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0155 - mse: 0.0155 - mae: 0.0977 - val_loss: 0.3002 - val_mse: 0.3002 - val_mae: 0.4990 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0164 - mse: 0.0164 - mae: 0.0993 - val_loss: 0.3001 - val_mse: 0.3001 - val_mae: 0.4989 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0161 - mse: 0.0161 - mae: 0.1021 - val_loss: 0.3002 - val_mse: 0.3002 - val_mae: 0.4990 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.24821
17/17 - 0s - loss: 0.0159 - mse: 0.0159 - mae: 0.0985 - val_loss: 0.3003 - val_mse: 0.3003 - val_mae: 0.4991 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 00052: early stopping
SMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 37.148636819682245
RMSE: 6.094968155756209
MAPE: 5.090179331518223
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 47.39% Accuracy
MSE: 31.582654654902484
RMSE: 5.619844718041815
MAPE: 4.507182634072088
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 65.00101872981564
RMSE: 8.062320926992156
MAPE: 6.705711592581163
DEMA
DEMA([input_arrays], [timeperiod=30])
Double Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
89
Working on DEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.774, Time=2.05 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=4.01 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15590.302, Time=7.02 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.43 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15269.503, Time=7.16 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-16414.961, Time=8.21 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16878.396, Time=9.90 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17030.019, Time=4.60 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.613, Time=3.06 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.441, Time=6.28 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=14.03 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17003.985, Time=3.33 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-16998.665, Time=3.36 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 78.451 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.721
Date: Sun, 12 Dec 2021 AIC -17087.441
Time: 15:16:00 BIC -16965.479
Sample: 0 HQIC -17040.603
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.799e-10 1.43e-20 -1.96e+10 0.000 -2.8e-10 -2.8e-10
x2 -2.817e-10 1.43e-20 -1.97e+10 0.000 -2.82e-10 -2.82e-10
x3 -2.805e-10 1.43e-20 -1.96e+10 0.000 -2.8e-10 -2.8e-10
x4 1.0000 1.43e-20 7e+19 0.000 1.000 1.000
x5 -2.597e-10 1.37e-20 -1.89e+10 0.000 -2.6e-10 -2.6e-10
x6 -1.388e-09 3.12e-20 -4.45e+10 0.000 -1.39e-09 -1.39e-09
x7 -2.789e-10 1.42e-20 -1.96e+10 0.000 -2.79e-10 -2.79e-10
x8 -2.76e-10 1.42e-20 -1.95e+10 0.000 -2.76e-10 -2.76e-10
x9 -2.216e-12 3.53e-22 -6.28e+09 0.000 -2.22e-12 -2.22e-12
x10 -1.345e-10 9.82e-21 -1.37e+10 0.000 -1.34e-10 -1.34e-10
x11 -2.898e-10 1.45e-20 -2e+10 0.000 -2.9e-10 -2.9e-10
x12 -2.602e-10 1.38e-20 -1.89e+10 0.000 -2.6e-10 -2.6e-10
x13 -2.807e-10 1.43e-20 -1.96e+10 0.000 -2.81e-10 -2.81e-10
x14 -1.87e-09 3.69e-20 -5.07e+10 0.000 -1.87e-09 -1.87e-09
x15 -2.726e-10 1.43e-20 -1.9e+10 0.000 -2.73e-10 -2.73e-10
x16 -7.915e-11 7.68e-21 -1.03e+10 0.000 -7.92e-11 -7.92e-11
x17 -2.606e-10 1.33e-20 -1.96e+10 0.000 -2.61e-10 -2.61e-10
x18 -6.408e-10 2.16e-20 -2.97e+10 0.000 -6.41e-10 -6.41e-10
x19 -2.881e-10 1.46e-20 -1.98e+10 0.000 -2.88e-10 -2.88e-10
x20 -4.337e-10 1.78e-20 -2.44e+10 0.000 -4.34e-10 -4.34e-10
x21 -4.549e-10 1.79e-20 -2.55e+10 0.000 -4.55e-10 -4.55e-10
ar.L1 -0.4923 1.46e-22 -3.38e+21 0.000 -0.492 -0.492
ar.L2 -0.1923 8.47e-23 -2.27e+21 0.000 -0.192 -0.192
ar.L3 -0.0461 4.02e-23 -1.15e+21 0.000 -0.046 -0.046
ma.L1 -0.7078 3.31e-22 -2.14e+21 0.000 -0.708 -0.708
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 55.07 Jarque-Bera (JB): 4171695.82
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.26
Prob(H) (two-sided): 0.00 Kurtosis: 355.51
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 2.62e+41. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.73965, saving model to LSTM7.h5
10/10 - 2s - loss: 0.1233 - mse: 0.1233 - mae: 0.2942 - val_loss: 0.7397 - val_mse: 0.7397 - val_mae: 0.8252 - lr: 0.0010 - 2s/epoch - 218ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.73965 to 0.66869, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0684 - mse: 0.0684 - mae: 0.2238 - val_loss: 0.6687 - val_mse: 0.6687 - val_mae: 0.7838 - lr: 0.0010 - 64ms/epoch - 6ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.66869 to 0.58974, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0340 - mse: 0.0340 - mae: 0.1484 - val_loss: 0.5897 - val_mse: 0.5897 - val_mae: 0.7351 - lr: 0.0010 - 62ms/epoch - 6ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.58974 to 0.42411, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0324 - mse: 0.0324 - mae: 0.1428 - val_loss: 0.4241 - val_mse: 0.4241 - val_mae: 0.6198 - lr: 0.0010 - 60ms/epoch - 6ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.42411 to 0.33734, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0242 - mse: 0.0242 - mae: 0.1266 - val_loss: 0.3373 - val_mse: 0.3373 - val_mae: 0.5500 - lr: 0.0010 - 65ms/epoch - 7ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.33734 to 0.30873, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0236 - mse: 0.0236 - mae: 0.1251 - val_loss: 0.3087 - val_mse: 0.3087 - val_mae: 0.5256 - lr: 0.0010 - 62ms/epoch - 6ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.30873 to 0.25836, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0202 - mse: 0.0202 - mae: 0.1135 - val_loss: 0.2584 - val_mse: 0.2584 - val_mae: 0.4789 - lr: 0.0010 - 62ms/epoch - 6ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.25836 to 0.25317, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0191 - mse: 0.0191 - mae: 0.1097 - val_loss: 0.2532 - val_mse: 0.2532 - val_mae: 0.4741 - lr: 0.0010 - 78ms/epoch - 8ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.25317
10/10 - 0s - loss: 0.0155 - mse: 0.0155 - mae: 0.0986 - val_loss: 0.2580 - val_mse: 0.2580 - val_mae: 0.4791 - lr: 0.0010 - 62ms/epoch - 6ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.25317 to 0.24856, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.0999 - val_loss: 0.2486 - val_mse: 0.2486 - val_mae: 0.4699 - lr: 0.0010 - 73ms/epoch - 7ms/step
Epoch 11/500
Epoch 00011: val_loss improved from 0.24856 to 0.24088, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0123 - mse: 0.0123 - mae: 0.0883 - val_loss: 0.2409 - val_mse: 0.2409 - val_mae: 0.4626 - lr: 0.0010 - 72ms/epoch - 7ms/step
Epoch 12/500
Epoch 00012: val_loss improved from 0.24088 to 0.21981, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0119 - mse: 0.0119 - mae: 0.0874 - val_loss: 0.2198 - val_mse: 0.2198 - val_mae: 0.4406 - lr: 0.0010 - 74ms/epoch - 7ms/step
Epoch 13/500
Epoch 00013: val_loss improved from 0.21981 to 0.20760, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0119 - mse: 0.0119 - mae: 0.0871 - val_loss: 0.2076 - val_mse: 0.2076 - val_mae: 0.4273 - lr: 0.0010 - 64ms/epoch - 6ms/step
Epoch 14/500
Epoch 00014: val_loss improved from 0.20760 to 0.20492, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0115 - mse: 0.0115 - mae: 0.0854 - val_loss: 0.2049 - val_mse: 0.2049 - val_mae: 0.4246 - lr: 0.0010 - 65ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss improved from 0.20492 to 0.20166, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0092 - mse: 0.0092 - mae: 0.0765 - val_loss: 0.2017 - val_mse: 0.2017 - val_mae: 0.4214 - lr: 0.0010 - 64ms/epoch - 6ms/step
Epoch 16/500
Epoch 00016: val_loss improved from 0.20166 to 0.19058, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0098 - mse: 0.0098 - mae: 0.0777 - val_loss: 0.1906 - val_mse: 0.1906 - val_mae: 0.4091 - lr: 0.0010 - 67ms/epoch - 7ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.19058
10/10 - 0s - loss: 0.0082 - mse: 0.0082 - mae: 0.0727 - val_loss: 0.1970 - val_mse: 0.1970 - val_mae: 0.4164 - lr: 0.0010 - 51ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.19058
10/10 - 0s - loss: 0.0089 - mse: 0.0089 - mae: 0.0740 - val_loss: 0.2024 - val_mse: 0.2024 - val_mae: 0.4221 - lr: 0.0010 - 52ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.19058
10/10 - 0s - loss: 0.0083 - mse: 0.0083 - mae: 0.0707 - val_loss: 0.2001 - val_mse: 0.2001 - val_mae: 0.4194 - lr: 0.0010 - 56ms/epoch - 6ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.19058
10/10 - 0s - loss: 0.0079 - mse: 0.0079 - mae: 0.0698 - val_loss: 0.2042 - val_mse: 0.2042 - val_mae: 0.4243 - lr: 0.0010 - 50ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss improved from 0.19058 to 0.18832, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0076 - mse: 0.0076 - mae: 0.0692 - val_loss: 0.1883 - val_mse: 0.1883 - val_mae: 0.4066 - lr: 0.0010 - 63ms/epoch - 6ms/step
Epoch 22/500
Epoch 00022: val_loss improved from 0.18832 to 0.16796, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0076 - mse: 0.0076 - mae: 0.0674 - val_loss: 0.1680 - val_mse: 0.1680 - val_mae: 0.3824 - lr: 0.0010 - 64ms/epoch - 6ms/step
Epoch 23/500
Epoch 00023: val_loss improved from 0.16796 to 0.16784, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0071 - mse: 0.0071 - mae: 0.0667 - val_loss: 0.1678 - val_mse: 0.1678 - val_mae: 0.3821 - lr: 0.0010 - 74ms/epoch - 7ms/step
Epoch 24/500
Epoch 00024: val_loss improved from 0.16784 to 0.16775, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0622 - val_loss: 0.1678 - val_mse: 0.1678 - val_mae: 0.3819 - lr: 0.0010 - 74ms/epoch - 7ms/step
Epoch 25/500
Epoch 00025: val_loss improved from 0.16775 to 0.15586, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0644 - val_loss: 0.1559 - val_mse: 0.1559 - val_mae: 0.3670 - lr: 0.0010 - 71ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.15586
10/10 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0616 - val_loss: 0.1584 - val_mse: 0.1584 - val_mae: 0.3704 - lr: 0.0010 - 55ms/epoch - 6ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.15586
10/10 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0617 - val_loss: 0.1619 - val_mse: 0.1619 - val_mae: 0.3742 - lr: 0.0010 - 52ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss improved from 0.15586 to 0.15265, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0602 - val_loss: 0.1526 - val_mse: 0.1526 - val_mae: 0.3623 - lr: 0.0010 - 68ms/epoch - 7ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.15265
10/10 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0586 - val_loss: 0.1528 - val_mse: 0.1528 - val_mae: 0.3626 - lr: 0.0010 - 53ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss improved from 0.15265 to 0.14845, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0573 - val_loss: 0.1484 - val_mse: 0.1484 - val_mae: 0.3568 - lr: 0.0010 - 68ms/epoch - 7ms/step
Epoch 31/500
Epoch 00031: val_loss improved from 0.14845 to 0.13058, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0053 - mse: 0.0053 - mae: 0.0577 - val_loss: 0.1306 - val_mse: 0.1306 - val_mae: 0.3325 - lr: 0.0010 - 69ms/epoch - 7ms/step
Epoch 32/500
Epoch 00032: val_loss improved from 0.13058 to 0.12945, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0569 - val_loss: 0.1295 - val_mse: 0.1295 - val_mae: 0.3306 - lr: 0.0010 - 67ms/epoch - 7ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.12945
10/10 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0541 - val_loss: 0.1421 - val_mse: 0.1421 - val_mae: 0.3480 - lr: 0.0010 - 53ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.12945
10/10 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0522 - val_loss: 0.1620 - val_mse: 0.1620 - val_mae: 0.3739 - lr: 0.0010 - 51ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.12945
10/10 - 0s - loss: 0.0050 - mse: 0.0050 - mae: 0.0551 - val_loss: 0.1601 - val_mse: 0.1601 - val_mae: 0.3718 - lr: 0.0010 - 57ms/epoch - 6ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.12945
10/10 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0564 - val_loss: 0.1506 - val_mse: 0.1506 - val_mae: 0.3599 - lr: 0.0010 - 55ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss improved from 0.12945 to 0.12770, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0531 - val_loss: 0.1277 - val_mse: 0.1277 - val_mae: 0.3292 - lr: 0.0010 - 65ms/epoch - 6ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.12770
10/10 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0518 - val_loss: 0.1278 - val_mse: 0.1278 - val_mae: 0.3294 - lr: 0.0010 - 63ms/epoch - 6ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.12770
10/10 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0500 - val_loss: 0.1368 - val_mse: 0.1368 - val_mae: 0.3418 - lr: 0.0010 - 62ms/epoch - 6ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.12770
10/10 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0542 - val_loss: 0.1507 - val_mse: 0.1507 - val_mae: 0.3599 - lr: 0.0010 - 58ms/epoch - 6ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.12770
10/10 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0523 - val_loss: 0.1399 - val_mse: 0.1399 - val_mae: 0.3455 - lr: 0.0010 - 60ms/epoch - 6ms/step
Epoch 42/500
Epoch 00042: val_loss improved from 0.12770 to 0.12669, saving model to LSTM7.h5
10/10 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0536 - val_loss: 0.1267 - val_mse: 0.1267 - val_mae: 0.3273 - lr: 0.0010 - 74ms/epoch - 7ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0488 - val_loss: 0.1534 - val_mse: 0.1534 - val_mae: 0.3634 - lr: 0.0010 - 61ms/epoch - 6ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0525 - val_loss: 0.1543 - val_mse: 0.1543 - val_mae: 0.3645 - lr: 0.0010 - 59ms/epoch - 6ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0493 - val_loss: 0.1419 - val_mse: 0.1419 - val_mae: 0.3487 - lr: 0.0010 - 53ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0500 - val_loss: 0.1270 - val_mse: 0.1270 - val_mae: 0.3286 - lr: 0.0010 - 58ms/epoch - 6ms/step
Epoch 47/500
Epoch 00047: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00047: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0519 - val_loss: 0.1277 - val_mse: 0.1277 - val_mae: 0.3294 - lr: 0.0010 - 53ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0489 - val_loss: 0.1309 - val_mse: 0.1309 - val_mae: 0.3338 - lr: 1.0000e-04 - 51ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0495 - val_loss: 0.1321 - val_mse: 0.1321 - val_mae: 0.3355 - lr: 1.0000e-04 - 51ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0496 - val_loss: 0.1329 - val_mse: 0.1329 - val_mae: 0.3364 - lr: 1.0000e-04 - 55ms/epoch - 6ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0491 - val_loss: 0.1356 - val_mse: 0.1356 - val_mae: 0.3400 - lr: 1.0000e-04 - 52ms/epoch - 5ms/step
Epoch 52/500
Epoch 00052: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00052: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0484 - val_loss: 0.1380 - val_mse: 0.1380 - val_mae: 0.3433 - lr: 1.0000e-04 - 52ms/epoch - 5ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0493 - val_loss: 0.1381 - val_mse: 0.1381 - val_mae: 0.3435 - lr: 1.0000e-05 - 55ms/epoch - 6ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0491 - val_loss: 0.1383 - val_mse: 0.1383 - val_mae: 0.3437 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0464 - val_loss: 0.1383 - val_mse: 0.1383 - val_mae: 0.3438 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0463 - val_loss: 0.1384 - val_mse: 0.1384 - val_mae: 0.3438 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 57/500
Epoch 00057: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00057: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0485 - val_loss: 0.1384 - val_mse: 0.1384 - val_mae: 0.3438 - lr: 1.0000e-05 - 55ms/epoch - 6ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0493 - val_loss: 0.1384 - val_mse: 0.1384 - val_mae: 0.3438 - lr: 1.0000e-05 - 57ms/epoch - 6ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0474 - val_loss: 0.1387 - val_mse: 0.1387 - val_mae: 0.3443 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0479 - val_loss: 0.1390 - val_mse: 0.1390 - val_mae: 0.3447 - lr: 1.0000e-05 - 52ms/epoch - 5ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0484 - val_loss: 0.1394 - val_mse: 0.1394 - val_mae: 0.3451 - lr: 1.0000e-05 - 53ms/epoch - 5ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0491 - val_loss: 0.1394 - val_mse: 0.1394 - val_mae: 0.3452 - lr: 1.0000e-05 - 53ms/epoch - 5ms/step
Epoch 63/500
Epoch 00063: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0486 - val_loss: 0.1392 - val_mse: 0.1392 - val_mae: 0.3450 - lr: 1.0000e-05 - 56ms/epoch - 6ms/step
Epoch 64/500
Epoch 00064: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0035 - mse: 0.0035 - mae: 0.0465 - val_loss: 0.1390 - val_mse: 0.1390 - val_mae: 0.3446 - lr: 1.0000e-05 - 57ms/epoch - 6ms/step
Epoch 65/500
Epoch 00065: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0035 - mse: 0.0035 - mae: 0.0467 - val_loss: 0.1390 - val_mse: 0.1390 - val_mae: 0.3446 - lr: 1.0000e-05 - 53ms/epoch - 5ms/step
Epoch 66/500
Epoch 00066: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0472 - val_loss: 0.1392 - val_mse: 0.1392 - val_mae: 0.3449 - lr: 1.0000e-05 - 52ms/epoch - 5ms/step
Epoch 67/500
Epoch 00067: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0493 - val_loss: 0.1392 - val_mse: 0.1392 - val_mae: 0.3449 - lr: 1.0000e-05 - 55ms/epoch - 6ms/step
Epoch 68/500
Epoch 00068: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0479 - val_loss: 0.1391 - val_mse: 0.1391 - val_mae: 0.3448 - lr: 1.0000e-05 - 53ms/epoch - 5ms/step
Epoch 69/500
Epoch 00069: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0492 - val_loss: 0.1392 - val_mse: 0.1392 - val_mae: 0.3449 - lr: 1.0000e-05 - 53ms/epoch - 5ms/step
Epoch 70/500
Epoch 00070: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0479 - val_loss: 0.1394 - val_mse: 0.1394 - val_mae: 0.3451 - lr: 1.0000e-05 - 55ms/epoch - 5ms/step
Epoch 71/500
Epoch 00071: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0479 - val_loss: 0.1394 - val_mse: 0.1394 - val_mae: 0.3452 - lr: 1.0000e-05 - 67ms/epoch - 7ms/step
Epoch 72/500
Epoch 00072: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0469 - val_loss: 0.1395 - val_mse: 0.1395 - val_mae: 0.3453 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 73/500
Epoch 00073: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0470 - val_loss: 0.1395 - val_mse: 0.1395 - val_mae: 0.3453 - lr: 1.0000e-05 - 57ms/epoch - 6ms/step
Epoch 74/500
Epoch 00074: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0470 - val_loss: 0.1393 - val_mse: 0.1393 - val_mae: 0.3451 - lr: 1.0000e-05 - 53ms/epoch - 5ms/step
Epoch 75/500
Epoch 00075: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0480 - val_loss: 0.1395 - val_mse: 0.1395 - val_mae: 0.3453 - lr: 1.0000e-05 - 56ms/epoch - 6ms/step
Epoch 76/500
Epoch 00076: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0475 - val_loss: 0.1395 - val_mse: 0.1395 - val_mae: 0.3453 - lr: 1.0000e-05 - 62ms/epoch - 6ms/step
Epoch 77/500
Epoch 00077: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0474 - val_loss: 0.1398 - val_mse: 0.1398 - val_mae: 0.3457 - lr: 1.0000e-05 - 51ms/epoch - 5ms/step
Epoch 78/500
Epoch 00078: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0484 - val_loss: 0.1400 - val_mse: 0.1400 - val_mae: 0.3460 - lr: 1.0000e-05 - 53ms/epoch - 5ms/step
Epoch 79/500
Epoch 00079: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0474 - val_loss: 0.1400 - val_mse: 0.1400 - val_mae: 0.3460 - lr: 1.0000e-05 - 53ms/epoch - 5ms/step
Epoch 80/500
Epoch 00080: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0468 - val_loss: 0.1399 - val_mse: 0.1399 - val_mae: 0.3459 - lr: 1.0000e-05 - 57ms/epoch - 6ms/step
Epoch 81/500
Epoch 00081: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0035 - mse: 0.0035 - mae: 0.0479 - val_loss: 0.1400 - val_mse: 0.1400 - val_mae: 0.3460 - lr: 1.0000e-05 - 51ms/epoch - 5ms/step
Epoch 82/500
Epoch 00082: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0489 - val_loss: 0.1399 - val_mse: 0.1399 - val_mae: 0.3459 - lr: 1.0000e-05 - 52ms/epoch - 5ms/step
Epoch 83/500
Epoch 00083: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0035 - mse: 0.0035 - mae: 0.0472 - val_loss: 0.1396 - val_mse: 0.1396 - val_mae: 0.3454 - lr: 1.0000e-05 - 55ms/epoch - 5ms/step
Epoch 84/500
Epoch 00084: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0496 - val_loss: 0.1394 - val_mse: 0.1394 - val_mae: 0.3451 - lr: 1.0000e-05 - 52ms/epoch - 5ms/step
Epoch 85/500
Epoch 00085: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0475 - val_loss: 0.1394 - val_mse: 0.1394 - val_mae: 0.3452 - lr: 1.0000e-05 - 53ms/epoch - 5ms/step
Epoch 86/500
Epoch 00086: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0035 - mse: 0.0035 - mae: 0.0470 - val_loss: 0.1397 - val_mse: 0.1397 - val_mae: 0.3455 - lr: 1.0000e-05 - 56ms/epoch - 6ms/step
Epoch 87/500
Epoch 00087: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0471 - val_loss: 0.1397 - val_mse: 0.1397 - val_mae: 0.3456 - lr: 1.0000e-05 - 55ms/epoch - 6ms/step
Epoch 88/500
Epoch 00088: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0484 - val_loss: 0.1396 - val_mse: 0.1396 - val_mae: 0.3454 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step
Epoch 89/500
Epoch 00089: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0482 - val_loss: 0.1394 - val_mse: 0.1394 - val_mae: 0.3452 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 90/500
Epoch 00090: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0506 - val_loss: 0.1396 - val_mse: 0.1396 - val_mae: 0.3454 - lr: 1.0000e-05 - 56ms/epoch - 6ms/step
Epoch 91/500
Epoch 00091: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0035 - mse: 0.0035 - mae: 0.0466 - val_loss: 0.1399 - val_mse: 0.1399 - val_mae: 0.3458 - lr: 1.0000e-05 - 55ms/epoch - 5ms/step
Epoch 92/500
Epoch 00092: val_loss did not improve from 0.12669
10/10 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0471 - val_loss: 0.1402 - val_mse: 0.1402 - val_mae: 0.3462 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 00092: early stopping
SMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 37.148636819682245
RMSE: 6.094968155756209
MAPE: 5.090179331518223
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 47.39% Accuracy
MSE: 31.582654654902484
RMSE: 5.619844718041815
MAPE: 4.507182634072088
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 65.00101872981564
RMSE: 8.062320926992156
MAPE: 6.705711592581163
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 35.269002386685244
RMSE: 5.938771117553298
MAPE: 4.62878838931535
KAMA
KAMA([input_arrays], [timeperiod=30])
Kaufman Adaptive Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
18
Working on KAMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.902, Time=2.34 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=3.99 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16796.316, Time=8.11 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=4.91 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-17004.193, Time=2.42 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15176.063, Time=10.20 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16873.638, Time=10.61 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17008.756, Time=2.40 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.764, Time=2.82 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-15723.849, Time=13.26 sec
ARIMA(2,3,0)(0,0,0)[0] intercept : AIC=-17006.756, Time=2.38 sec
Best model: ARIMA(2,3,0)(0,0,0)[0]
Total fit time: 63.483 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(2, 3, 0) Log Likelihood 8528.378
Date: Sun, 12 Dec 2021 AIC -17008.756
Time: 15:24:53 BIC -16896.176
Sample: 0 HQIC -16965.520
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.24e-15 7.41e-26 -3.02e+10 0.000 -2.24e-15 -2.24e-15
x2 8.461e-16 6.6e-26 1.28e+10 0.000 8.46e-16 8.46e-16
x3 4.901e-16 6.89e-26 7.11e+09 0.000 4.9e-16 4.9e-16
x4 1.0000 6.96e-26 1.44e+25 0.000 1.000 1.000
x5 5.931e-15 6.61e-26 8.97e+10 0.000 5.93e-15 5.93e-15
x6 -1.05e-15 1.5e-25 -7e+09 0.000 -1.05e-15 -1.05e-15
x7 1.439e-15 6.87e-26 2.1e+10 0.000 1.44e-15 1.44e-15
x8 -1.25e-15 6.8e-26 -1.84e+10 0.000 -1.25e-15 -1.25e-15
x9 -9.356e-17 8.97e-27 -1.04e+10 0.000 -9.36e-17 -9.36e-17
x10 -1.145e-16 2.88e-26 -3.98e+09 0.000 -1.15e-16 -1.15e-16
x11 -2.036e-16 6.8e-26 -3e+09 0.000 -2.04e-16 -2.04e-16
x12 5.951e-16 6.76e-26 8.81e+09 0.000 5.95e-16 5.95e-16
x13 -6.117e-17 6.94e-26 -8.81e+08 0.000 -6.12e-17 -6.12e-17
x14 1.167e-15 1.99e-25 5.85e+09 0.000 1.17e-15 1.17e-15
x15 -4.274e-14 6.99e-26 -6.11e+11 0.000 -4.27e-14 -4.27e-14
x16 2.262e-14 8.56e-26 2.64e+11 0.000 2.26e-14 2.26e-14
x17 3.384e-14 6.46e-26 5.24e+11 0.000 3.38e-14 3.38e-14
x18 9.894e-16 5.8e-26 1.71e+10 0.000 9.89e-16 9.89e-16
x19 4.115e-14 7.75e-26 5.31e+11 0.000 4.12e-14 4.12e-14
x20 -2.176e-15 9.49e-26 -2.29e+10 0.000 -2.18e-15 -2.18e-15
x21 -7.755e-17 4.63e-26 -1.67e+09 0.000 -7.75e-17 -7.75e-17
ar.L1 -0.9988 9.76e-22 -1.02e+21 0.000 -0.999 -0.999
ar.L2 -0.4972 4.07e-23 -1.22e+22 0.000 -0.497 -0.497
sigma2 1e-10 6.99e-11 1.432 0.152 -3.69e-11 2.37e-10
===================================================================================
Ljung-Box (L1) (Q): 31.54 Jarque-Bera (JB): 2432532.03
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: -0.15
Prob(H) (two-sided): 0.00 Kurtosis: 272.30
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 7.19e+48. Standard errors may be unstable.
ARIMA order: (2, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.01543, saving model to LSTM7.h5
45/45 - 3s - loss: 0.0923 - mse: 0.0923 - mae: 0.2595 - val_loss: 0.0154 - val_mse: 0.0154 - val_mae: 0.1058 - lr: 0.0010 - 3s/epoch - 57ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.01543
45/45 - 0s - loss: 0.0162 - mse: 0.0162 - mae: 0.1027 - val_loss: 0.0382 - val_mse: 0.0382 - val_mae: 0.1807 - lr: 0.0010 - 198ms/epoch - 4ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.01543
45/45 - 0s - loss: 0.0192 - mse: 0.0192 - mae: 0.1081 - val_loss: 0.0306 - val_mse: 0.0306 - val_mae: 0.1598 - lr: 0.0010 - 172ms/epoch - 4ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.01543
45/45 - 0s - loss: 0.0143 - mse: 0.0143 - mae: 0.0924 - val_loss: 0.0376 - val_mse: 0.0376 - val_mae: 0.1802 - lr: 0.0010 - 173ms/epoch - 4ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.01543
45/45 - 0s - loss: 0.0151 - mse: 0.0151 - mae: 0.0952 - val_loss: 0.0269 - val_mse: 0.0269 - val_mae: 0.1482 - lr: 0.0010 - 179ms/epoch - 4ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.01543
45/45 - 0s - loss: 0.0145 - mse: 0.0145 - mae: 0.0929 - val_loss: 0.0243 - val_mse: 0.0243 - val_mae: 0.1400 - lr: 0.0010 - 175ms/epoch - 4ms/step
Epoch 7/500
Epoch 00007: val_loss improved from 0.01543 to 0.00857, saving model to LSTM7.h5
45/45 - 0s - loss: 0.0341 - mse: 0.0341 - mae: 0.1510 - val_loss: 0.0086 - val_mse: 0.0086 - val_mae: 0.0754 - lr: 1.0000e-04 - 198ms/epoch - 4ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.00857 to 0.00771, saving model to LSTM7.h5
45/45 - 0s - loss: 0.0104 - mse: 0.0104 - mae: 0.0835 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0711 - lr: 1.0000e-04 - 194ms/epoch - 4ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.00771 to 0.00749, saving model to LSTM7.h5
45/45 - 0s - loss: 0.0093 - mse: 0.0093 - mae: 0.0781 - val_loss: 0.0075 - val_mse: 0.0075 - val_mae: 0.0701 - lr: 1.0000e-04 - 180ms/epoch - 4ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.00749 to 0.00737, saving model to LSTM7.h5
45/45 - 0s - loss: 0.0087 - mse: 0.0087 - mae: 0.0755 - val_loss: 0.0074 - val_mse: 0.0074 - val_mae: 0.0695 - lr: 1.0000e-04 - 209ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: val_loss improved from 0.00737 to 0.00729, saving model to LSTM7.h5
45/45 - 0s - loss: 0.0087 - mse: 0.0087 - mae: 0.0747 - val_loss: 0.0073 - val_mse: 0.0073 - val_mae: 0.0691 - lr: 1.0000e-04 - 179ms/epoch - 4ms/step
Epoch 12/500
Epoch 00012: val_loss improved from 0.00729 to 0.00724, saving model to LSTM7.h5
45/45 - 0s - loss: 0.0086 - mse: 0.0086 - mae: 0.0747 - val_loss: 0.0072 - val_mse: 0.0072 - val_mae: 0.0688 - lr: 1.0000e-04 - 211ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0079 - mse: 0.0079 - mae: 0.0719 - val_loss: 0.0075 - val_mse: 0.0075 - val_mae: 0.0700 - lr: 1.0000e-04 - 183ms/epoch - 4ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0700 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0716 - lr: 1.0000e-04 - 177ms/epoch - 4ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0085 - mse: 0.0085 - mae: 0.0741 - val_loss: 0.0081 - val_mse: 0.0081 - val_mae: 0.0735 - lr: 1.0000e-04 - 171ms/epoch - 4ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0706 - val_loss: 0.0085 - val_mse: 0.0085 - val_mae: 0.0759 - lr: 1.0000e-04 - 176ms/epoch - 4ms/step
Epoch 17/500
Epoch 00017: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00017: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0076 - mse: 0.0076 - mae: 0.0706 - val_loss: 0.0081 - val_mse: 0.0081 - val_mae: 0.0738 - lr: 1.0000e-04 - 183ms/epoch - 4ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0635 - val_loss: 0.0080 - val_mse: 0.0080 - val_mae: 0.0733 - lr: 1.0000e-05 - 187ms/epoch - 4ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0659 - val_loss: 0.0080 - val_mse: 0.0080 - val_mae: 0.0732 - lr: 1.0000e-05 - 168ms/epoch - 4ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0646 - val_loss: 0.0080 - val_mse: 0.0080 - val_mae: 0.0733 - lr: 1.0000e-05 - 170ms/epoch - 4ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0638 - val_loss: 0.0080 - val_mse: 0.0080 - val_mae: 0.0732 - lr: 1.0000e-05 - 182ms/epoch - 4ms/step
Epoch 22/500
Epoch 00022: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00022: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0071 - mse: 0.0071 - mae: 0.0671 - val_loss: 0.0079 - val_mse: 0.0079 - val_mae: 0.0728 - lr: 1.0000e-05 - 169ms/epoch - 4ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0653 - val_loss: 0.0079 - val_mse: 0.0079 - val_mae: 0.0727 - lr: 1.0000e-05 - 198ms/epoch - 4ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0661 - val_loss: 0.0078 - val_mse: 0.0078 - val_mae: 0.0724 - lr: 1.0000e-05 - 179ms/epoch - 4ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0071 - mse: 0.0071 - mae: 0.0658 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0721 - lr: 1.0000e-05 - 177ms/epoch - 4ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0642 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0719 - lr: 1.0000e-05 - 177ms/epoch - 4ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0641 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0716 - lr: 1.0000e-05 - 167ms/epoch - 4ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0650 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0710 - lr: 1.0000e-05 - 182ms/epoch - 4ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0639 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0710 - lr: 1.0000e-05 - 181ms/epoch - 4ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0630 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0714 - lr: 1.0000e-05 - 169ms/epoch - 4ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0588 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0714 - lr: 1.0000e-05 - 167ms/epoch - 4ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0623 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0714 - lr: 1.0000e-05 - 179ms/epoch - 4ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0641 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0715 - lr: 1.0000e-05 - 167ms/epoch - 4ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0623 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0717 - lr: 1.0000e-05 - 194ms/epoch - 4ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0618 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0712 - lr: 1.0000e-05 - 185ms/epoch - 4ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0637 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0712 - lr: 1.0000e-05 - 173ms/epoch - 4ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0610 - val_loss: 0.0075 - val_mse: 0.0075 - val_mae: 0.0710 - lr: 1.0000e-05 - 176ms/epoch - 4ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0653 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0714 - lr: 1.0000e-05 - 174ms/epoch - 4ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0635 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0715 - lr: 1.0000e-05 - 179ms/epoch - 4ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0611 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0713 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0642 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0714 - lr: 1.0000e-05 - 168ms/epoch - 4ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0643 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0713 - lr: 1.0000e-05 - 169ms/epoch - 4ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0604 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0711 - lr: 1.0000e-05 - 167ms/epoch - 4ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0605 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0712 - lr: 1.0000e-05 - 184ms/epoch - 4ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0650 - val_loss: 0.0075 - val_mse: 0.0075 - val_mae: 0.0708 - lr: 1.0000e-05 - 193ms/epoch - 4ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0652 - val_loss: 0.0075 - val_mse: 0.0075 - val_mae: 0.0710 - lr: 1.0000e-05 - 181ms/epoch - 4ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0620 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0711 - lr: 1.0000e-05 - 173ms/epoch - 4ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0603 - val_loss: 0.0075 - val_mse: 0.0075 - val_mae: 0.0711 - lr: 1.0000e-05 - 166ms/epoch - 4ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0632 - val_loss: 0.0075 - val_mse: 0.0075 - val_mae: 0.0709 - lr: 1.0000e-05 - 184ms/epoch - 4ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0631 - val_loss: 0.0075 - val_mse: 0.0075 - val_mae: 0.0709 - lr: 1.0000e-05 - 190ms/epoch - 4ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0656 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0712 - lr: 1.0000e-05 - 175ms/epoch - 4ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0630 - val_loss: 0.0076 - val_mse: 0.0076 - val_mae: 0.0713 - lr: 1.0000e-05 - 170ms/epoch - 4ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0609 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0717 - lr: 1.0000e-05 - 165ms/epoch - 4ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0615 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0719 - lr: 1.0000e-05 - 181ms/epoch - 4ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0610 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0719 - lr: 1.0000e-05 - 170ms/epoch - 4ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0595 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0718 - lr: 1.0000e-05 - 187ms/epoch - 4ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0637 - val_loss: 0.0075 - val_mse: 0.0075 - val_mae: 0.0711 - lr: 1.0000e-05 - 179ms/epoch - 4ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0600 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0720 - lr: 1.0000e-05 - 165ms/epoch - 4ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0618 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0723 - lr: 1.0000e-05 - 174ms/epoch - 4ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0622 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0721 - lr: 1.0000e-05 - 174ms/epoch - 4ms/step
Epoch 61/500
Epoch 00061: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0580 - val_loss: 0.0079 - val_mse: 0.0079 - val_mae: 0.0732 - lr: 1.0000e-05 - 194ms/epoch - 4ms/step
Epoch 62/500
Epoch 00062: val_loss did not improve from 0.00724
45/45 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0606 - val_loss: 0.0077 - val_mse: 0.0077 - val_mae: 0.0721 - lr: 1.0000e-05 - 174ms/epoch - 4ms/step
Epoch 00062: early stopping
SMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 37.148636819682245
RMSE: 6.094968155756209
MAPE: 5.090179331518223
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 47.39% Accuracy
MSE: 31.582654654902484
RMSE: 5.619844718041815
MAPE: 4.507182634072088
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 65.00101872981564
RMSE: 8.062320926992156
MAPE: 6.705711592581163
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 35.269002386685244
RMSE: 5.938771117553298
MAPE: 4.62878838931535
KAMA
Prediction vs Close: 54.1% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 62.25156682504816
RMSE: 7.8899662119078915
MAPE: 6.222956717810362
MIDPOINT
MIDPOINT([input_arrays], [timeperiod=14])
MidPoint over period (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 14
Outputs:
real
14
Working on MIDPOINT predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.753, Time=2.21 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.592, Time=3.96 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16288.639, Time=11.12 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.592, Time=5.33 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15275.254, Time=7.23 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15486.751, Time=12.59 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=48.000, Time=0.49 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17008.491, Time=2.60 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.554, Time=3.16 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.445, Time=6.13 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=-15686.421, Time=9.69 sec
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17030.168, Time=15.39 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-15138.715, Time=14.21 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 94.129 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.722
Date: Sun, 12 Dec 2021 AIC -17087.445
Time: 15:27:32 BIC -16965.483
Sample: 0 HQIC -17040.607
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.14e-10 1.09e-20 -1.96e+10 0.000 -2.14e-10 -2.14e-10
x2 -2.126e-10 1.13e-20 -1.88e+10 0.000 -2.13e-10 -2.13e-10
x3 -2.175e-10 1.06e-20 -2.04e+10 0.000 -2.17e-10 -2.17e-10
x4 1.0000 1.1e-20 9.11e+19 0.000 1.000 1.000
x5 -1.941e-10 1.05e-20 -1.86e+10 0.000 -1.94e-10 -1.94e-10
x6 -4.131e-09 7.64e-20 -5.4e+10 0.000 -4.13e-09 -4.13e-09
x7 -1.965e-10 1.05e-20 -1.86e+10 0.000 -1.96e-10 -1.96e-10
x8 -1.961e-10 1.07e-20 -1.84e+10 0.000 -1.96e-10 -1.96e-10
x9 -1.005e-10 9.12e-22 -1.1e+11 0.000 -1e-10 -1e-10
x10 -1.739e-10 3.37e-21 -5.16e+10 0.000 -1.74e-10 -1.74e-10
x11 -1.941e-10 1.07e-20 -1.82e+10 0.000 -1.94e-10 -1.94e-10
x12 -2.005e-10 1.06e-20 -1.89e+10 0.000 -2e-10 -2e-10
x13 -2.056e-10 1.07e-20 -1.91e+10 0.000 -2.06e-10 -2.06e-10
x14 -1.687e-09 3.15e-20 -5.36e+10 0.000 -1.69e-09 -1.69e-09
x15 -2.365e-10 1.17e-20 -2.01e+10 0.000 -2.36e-10 -2.36e-10
x16 -1.523e-10 9.42e-21 -1.62e+10 0.000 -1.52e-10 -1.52e-10
x17 -1.491e-10 9.33e-21 -1.6e+10 0.000 -1.49e-10 -1.49e-10
x18 -6.404e-10 1.93e-20 -3.32e+10 0.000 -6.4e-10 -6.4e-10
x19 -2.596e-10 1.23e-20 -2.11e+10 0.000 -2.6e-10 -2.6e-10
x20 -6.246e-10 1.91e-20 -3.28e+10 0.000 -6.25e-10 -6.25e-10
x21 -1.953e-09 2.16e-20 -9.04e+10 0.000 -1.95e-09 -1.95e-09
ar.L1 -0.4914 1.46e-22 -3.38e+21 0.000 -0.491 -0.491
ar.L2 -0.1934 8.48e-23 -2.28e+21 0.000 -0.193 -0.193
ar.L3 -0.0491 4.2e-23 -1.17e+21 0.000 -0.049 -0.049
ma.L1 -0.7092 3.33e-22 -2.13e+21 0.000 -0.709 -0.709
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 32.51 Jarque-Bera (JB): 49038.38
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.35 Skew: 1.06
Prob(H) (two-sided): 0.00 Kurtosis: 41.18
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.71e+40. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.01564, saving model to LSTM7.h5
58/58 - 2s - loss: 0.1640 - mse: 0.1640 - mae: 0.2979 - val_loss: 0.0156 - val_mse: 0.0156 - val_mae: 0.1028 - lr: 0.0010 - 2s/epoch - 40ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.01564 to 0.01327, saving model to LSTM7.h5
58/58 - 0s - loss: 0.0291 - mse: 0.0291 - mae: 0.1345 - val_loss: 0.0133 - val_mse: 0.0133 - val_mae: 0.0936 - lr: 0.0010 - 236ms/epoch - 4ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.01327
58/58 - 0s - loss: 0.0198 - mse: 0.0198 - mae: 0.1098 - val_loss: 0.0155 - val_mse: 0.0155 - val_mae: 0.1041 - lr: 0.0010 - 208ms/epoch - 4ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.01327 to 0.01189, saving model to LSTM7.h5
58/58 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0957 - val_loss: 0.0119 - val_mse: 0.0119 - val_mae: 0.0905 - lr: 0.0010 - 235ms/epoch - 4ms/step
Epoch 5/500
Epoch 00005: val_loss improved from 0.01189 to 0.01096, saving model to LSTM7.h5
58/58 - 0s - loss: 0.0123 - mse: 0.0123 - mae: 0.0869 - val_loss: 0.0110 - val_mse: 0.0110 - val_mae: 0.0869 - lr: 0.0010 - 236ms/epoch - 4ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.01096 to 0.00941, saving model to LSTM7.h5
58/58 - 0s - loss: 0.0106 - mse: 0.0106 - mae: 0.0801 - val_loss: 0.0094 - val_mse: 0.0094 - val_mae: 0.0801 - lr: 0.0010 - 224ms/epoch - 4ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.00941
58/58 - 0s - loss: 0.0111 - mse: 0.0111 - mae: 0.0821 - val_loss: 0.0098 - val_mse: 0.0098 - val_mae: 0.0820 - lr: 0.0010 - 208ms/epoch - 4ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.00941 to 0.00874, saving model to LSTM7.h5
58/58 - 0s - loss: 0.0098 - mse: 0.0098 - mae: 0.0761 - val_loss: 0.0087 - val_mse: 0.0087 - val_mae: 0.0768 - lr: 0.0010 - 239ms/epoch - 4ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00874
58/58 - 0s - loss: 0.0105 - mse: 0.0105 - mae: 0.0793 - val_loss: 0.0093 - val_mse: 0.0093 - val_mae: 0.0803 - lr: 0.0010 - 228ms/epoch - 4ms/step
Epoch 10/500
Epoch 00010: val_loss improved from 0.00874 to 0.00850, saving model to LSTM7.h5
58/58 - 0s - loss: 0.0103 - mse: 0.0103 - mae: 0.0764 - val_loss: 0.0085 - val_mse: 0.0085 - val_mae: 0.0750 - lr: 0.0010 - 246ms/epoch - 4ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0089 - mse: 0.0089 - mae: 0.0728 - val_loss: 0.0091 - val_mse: 0.0091 - val_mae: 0.0794 - lr: 0.0010 - 213ms/epoch - 4ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0097 - mse: 0.0097 - mae: 0.0739 - val_loss: 0.0088 - val_mse: 0.0088 - val_mae: 0.0768 - lr: 0.0010 - 217ms/epoch - 4ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0671 - val_loss: 0.0085 - val_mse: 0.0085 - val_mae: 0.0750 - lr: 0.0010 - 228ms/epoch - 4ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0109 - mse: 0.0109 - mae: 0.0778 - val_loss: 0.0085 - val_mse: 0.0085 - val_mae: 0.0747 - lr: 0.0010 - 217ms/epoch - 4ms/step
Epoch 15/500
Epoch 00015: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00015: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0071 - mse: 0.0071 - mae: 0.0636 - val_loss: 0.0094 - val_mse: 0.0094 - val_mae: 0.0767 - lr: 0.0010 - 232ms/epoch - 4ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0244 - mse: 0.0244 - mae: 0.1335 - val_loss: 0.0119 - val_mse: 0.0119 - val_mae: 0.0916 - lr: 1.0000e-04 - 221ms/epoch - 4ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0087 - mse: 0.0087 - mae: 0.0770 - val_loss: 0.0116 - val_mse: 0.0116 - val_mae: 0.0909 - lr: 1.0000e-04 - 209ms/epoch - 4ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0088 - mse: 0.0088 - mae: 0.0765 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0906 - lr: 1.0000e-04 - 227ms/epoch - 4ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0083 - mse: 0.0083 - mae: 0.0743 - val_loss: 0.0110 - val_mse: 0.0110 - val_mae: 0.0897 - lr: 1.0000e-04 - 212ms/epoch - 4ms/step
Epoch 20/500
Epoch 00020: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00020: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0077 - mse: 0.0077 - mae: 0.0714 - val_loss: 0.0110 - val_mse: 0.0110 - val_mae: 0.0899 - lr: 1.0000e-04 - 224ms/epoch - 4ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0629 - val_loss: 0.0111 - val_mse: 0.0111 - val_mae: 0.0903 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0632 - val_loss: 0.0112 - val_mse: 0.0112 - val_mae: 0.0909 - lr: 1.0000e-05 - 227ms/epoch - 4ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0661 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0914 - lr: 1.0000e-05 - 236ms/epoch - 4ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0635 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0916 - lr: 1.0000e-05 - 231ms/epoch - 4ms/step
Epoch 25/500
Epoch 00025: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00025: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0655 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0917 - lr: 1.0000e-05 - 218ms/epoch - 4ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0636 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0918 - lr: 1.0000e-05 - 225ms/epoch - 4ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0621 - val_loss: 0.0115 - val_mse: 0.0115 - val_mae: 0.0918 - lr: 1.0000e-05 - 225ms/epoch - 4ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0641 - val_loss: 0.0115 - val_mse: 0.0115 - val_mae: 0.0921 - lr: 1.0000e-05 - 226ms/epoch - 4ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0610 - val_loss: 0.0115 - val_mse: 0.0115 - val_mae: 0.0921 - lr: 1.0000e-05 - 226ms/epoch - 4ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0606 - val_loss: 0.0115 - val_mse: 0.0115 - val_mae: 0.0920 - lr: 1.0000e-05 - 238ms/epoch - 4ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0640 - val_loss: 0.0115 - val_mse: 0.0115 - val_mae: 0.0920 - lr: 1.0000e-05 - 221ms/epoch - 4ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0606 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0919 - lr: 1.0000e-05 - 217ms/epoch - 4ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0601 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0919 - lr: 1.0000e-05 - 221ms/epoch - 4ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0623 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0920 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0627 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0920 - lr: 1.0000e-05 - 225ms/epoch - 4ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0632 - val_loss: 0.0115 - val_mse: 0.0115 - val_mae: 0.0921 - lr: 1.0000e-05 - 229ms/epoch - 4ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0609 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0919 - lr: 1.0000e-05 - 227ms/epoch - 4ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0634 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0920 - lr: 1.0000e-05 - 219ms/epoch - 4ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0627 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0920 - lr: 1.0000e-05 - 217ms/epoch - 4ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0605 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0919 - lr: 1.0000e-05 - 230ms/epoch - 4ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0635 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0917 - lr: 1.0000e-05 - 219ms/epoch - 4ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0605 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0917 - lr: 1.0000e-05 - 227ms/epoch - 4ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0660 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0919 - lr: 1.0000e-05 - 221ms/epoch - 4ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0605 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0918 - lr: 1.0000e-05 - 233ms/epoch - 4ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0600 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0917 - lr: 1.0000e-05 - 229ms/epoch - 4ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0615 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0919 - lr: 1.0000e-05 - 218ms/epoch - 4ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0619 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0918 - lr: 1.0000e-05 - 217ms/epoch - 4ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0617 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0919 - lr: 1.0000e-05 - 229ms/epoch - 4ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0612 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0920 - lr: 1.0000e-05 - 220ms/epoch - 4ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0597 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0922 - lr: 1.0000e-05 - 219ms/epoch - 4ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0617 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0923 - lr: 1.0000e-05 - 231ms/epoch - 4ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0580 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0924 - lr: 1.0000e-05 - 227ms/epoch - 4ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0593 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0923 - lr: 1.0000e-05 - 229ms/epoch - 4ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0602 - val_loss: 0.0114 - val_mse: 0.0114 - val_mae: 0.0922 - lr: 1.0000e-05 - 220ms/epoch - 4ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0589 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0919 - lr: 1.0000e-05 - 231ms/epoch - 4ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0579 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0920 - lr: 1.0000e-05 - 220ms/epoch - 4ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0604 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0920 - lr: 1.0000e-05 - 227ms/epoch - 4ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0607 - val_loss: 0.0113 - val_mse: 0.0113 - val_mae: 0.0917 - lr: 1.0000e-05 - 219ms/epoch - 4ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0617 - val_loss: 0.0112 - val_mse: 0.0112 - val_mae: 0.0914 - lr: 1.0000e-05 - 218ms/epoch - 4ms/step
Epoch 60/500
Epoch 00060: val_loss did not improve from 0.00850
58/58 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0599 - val_loss: 0.0112 - val_mse: 0.0112 - val_mae: 0.0914 - lr: 1.0000e-05 - 222ms/epoch - 4ms/step
Epoch 00060: early stopping
SMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 37.148636819682245
RMSE: 6.094968155756209
MAPE: 5.090179331518223
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 47.39% Accuracy
MSE: 31.582654654902484
RMSE: 5.619844718041815
MAPE: 4.507182634072088
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 65.00101872981564
RMSE: 8.062320926992156
MAPE: 6.705711592581163
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 35.269002386685244
RMSE: 5.938771117553298
MAPE: 4.62878838931535
KAMA
Prediction vs Close: 54.1% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 62.25156682504816
RMSE: 7.8899662119078915
MAPE: 6.222956717810362
MIDPOINT
Prediction vs Close: 51.12% Accuracy
Prediction vs Prediction: 42.54% Accuracy
MSE: 84.47531990876952
RMSE: 9.191045637399997
MAPE: 7.890202393641488
T3
T3([input_arrays], [timeperiod=5], [vfactor=0.7])
Triple Exponential Moving Average (T3) (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 5
vfactor: 0.7
Outputs:
real
19
Working on T3 predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-16569.270, Time=2.35 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14511.291, Time=2.49 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15408.738, Time=7.96 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-15165.005, Time=7.92 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15595.465, Time=6.94 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15837.470, Time=10.15 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-15491.538, Time=9.59 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-16378.438, Time=2.50 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-16318.604, Time=3.52 sec
ARIMA(1,3,1)(0,0,0)[0] intercept : AIC=-16567.270, Time=2.23 sec
Best model: ARIMA(1,3,1)(0,0,0)[0]
Total fit time: 55.672 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(1, 3, 1) Log Likelihood 8308.635
Date: Sun, 12 Dec 2021 AIC -16569.270
Time: 15:36:26 BIC -16456.690
Sample: 0 HQIC -16526.035
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 1.355e-13 3.43e-05 3.95e-09 1.000 -6.72e-05 6.72e-05
x2 5.009e-14 2.67e-05 1.88e-09 1.000 -5.23e-05 5.23e-05
x3 -9.101e-15 2.19e-05 -4.16e-10 1.000 -4.29e-05 4.29e-05
x4 1.0000 2.97e-05 3.37e+04 0.000 1.000 1.000
x5 3.626e-12 3.2e-05 1.13e-07 1.000 -6.28e-05 6.28e-05
x6 6.879e-17 0.000 5.13e-13 1.000 -0.000 0.000
x7 1.588e-13 4.04e-05 3.93e-09 1.000 -7.92e-05 7.92e-05
x8 -0.0002 9.77e-06 -20.395 0.000 -0.000 -0.000
x9 3.877e-14 0.001 6.24e-11 1.000 -0.001 0.001
x10 -7.41e-05 0.001 -0.129 0.897 -0.001 0.001
x11 0.0003 4.91e-05 6.926 0.000 0.000 0.000
x12 -0.0004 7.27e-05 -5.556 0.000 -0.001 -0.000
x13 -2.679e-14 3.39e-05 -7.9e-10 1.000 -6.65e-05 6.65e-05
x14 2.97e-13 0.000 2.31e-09 1.000 -0.000 0.000
x15 1.602e-12 7.47e-05 2.14e-08 1.000 -0.000 0.000
x16 -8.756e-13 4.29e-05 -2.04e-08 1.000 -8.41e-05 8.41e-05
x17 1.793e-12 6.56e-05 2.74e-08 1.000 -0.000 0.000
x18 -1.019e-13 0.000 -5.54e-10 1.000 -0.000 0.000
x19 -1.077e-12 8.29e-05 -1.3e-08 1.000 -0.000 0.000
x20 1.771e-13 8.45e-05 2.1e-09 1.000 -0.000 0.000
x21 9.233e-16 0.000 1.94e-12 1.000 -0.001 0.001
ar.L1 -0.2857 0.000 -2747.572 0.000 -0.286 -0.285
ma.L1 -0.9142 7.12e-06 -1.28e+05 0.000 -0.914 -0.914
sigma2 1e-10 7e-11 1.429 0.153 -3.71e-11 2.37e-10
===================================================================================
Ljung-Box (L1) (Q): 84.32 Jarque-Bera (JB): 4804295.53
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.87
Prob(H) (two-sided): 0.00 Kurtosis: 381.28
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 2.06e+20. Standard errors may be unstable.
ARIMA order: (1, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.06885, saving model to LSTM7.h5
43/43 - 2s - loss: 0.0587 - mse: 0.0587 - mae: 0.1834 - val_loss: 0.0689 - val_mse: 0.0689 - val_mae: 0.1965 - lr: 0.0010 - 2s/epoch - 52ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.06885 to 0.05274, saving model to LSTM7.h5
43/43 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1149 - val_loss: 0.0527 - val_mse: 0.0527 - val_mae: 0.1716 - lr: 0.0010 - 171ms/epoch - 4ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.05274
43/43 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1128 - val_loss: 0.0567 - val_mse: 0.0567 - val_mae: 0.1724 - lr: 0.0010 - 171ms/epoch - 4ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.05274
43/43 - 0s - loss: 0.0199 - mse: 0.0199 - mae: 0.1114 - val_loss: 0.0581 - val_mse: 0.0581 - val_mae: 0.1710 - lr: 0.0010 - 173ms/epoch - 4ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.05274
43/43 - 0s - loss: 0.0192 - mse: 0.0192 - mae: 0.1094 - val_loss: 0.0548 - val_mse: 0.0548 - val_mae: 0.1643 - lr: 0.0010 - 180ms/epoch - 4ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.05274
43/43 - 0s - loss: 0.0176 - mse: 0.0176 - mae: 0.1045 - val_loss: 0.0582 - val_mse: 0.0582 - val_mae: 0.1683 - lr: 0.0010 - 164ms/epoch - 4ms/step
Epoch 7/500
Epoch 00007: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00007: val_loss did not improve from 0.05274
43/43 - 0s - loss: 0.0187 - mse: 0.0187 - mae: 0.1111 - val_loss: 0.0579 - val_mse: 0.0579 - val_mae: 0.1680 - lr: 0.0010 - 156ms/epoch - 4ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.05274 to 0.03618, saving model to LSTM7.h5
43/43 - 0s - loss: 0.0386 - mse: 0.0386 - mae: 0.1606 - val_loss: 0.0362 - val_mse: 0.0362 - val_mae: 0.1372 - lr: 1.0000e-04 - 178ms/epoch - 4ms/step
Epoch 9/500
Epoch 00009: val_loss improved from 0.03618 to 0.03334, saving model to LSTM7.h5
43/43 - 0s - loss: 0.0114 - mse: 0.0114 - mae: 0.0866 - val_loss: 0.0333 - val_mse: 0.0333 - val_mae: 0.1355 - lr: 1.0000e-04 - 179ms/epoch - 4ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0089 - mse: 0.0089 - mae: 0.0771 - val_loss: 0.0336 - val_mse: 0.0336 - val_mae: 0.1359 - lr: 1.0000e-04 - 181ms/epoch - 4ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0085 - mse: 0.0085 - mae: 0.0752 - val_loss: 0.0352 - val_mse: 0.0352 - val_mae: 0.1373 - lr: 1.0000e-04 - 170ms/epoch - 4ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0728 - val_loss: 0.0366 - val_mse: 0.0366 - val_mae: 0.1391 - lr: 1.0000e-04 - 169ms/epoch - 4ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0079 - mse: 0.0079 - mae: 0.0714 - val_loss: 0.0378 - val_mse: 0.0378 - val_mae: 0.1407 - lr: 1.0000e-04 - 170ms/epoch - 4ms/step
Epoch 14/500
Epoch 00014: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00014: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0077 - mse: 0.0077 - mae: 0.0706 - val_loss: 0.0387 - val_mse: 0.0387 - val_mae: 0.1423 - lr: 1.0000e-04 - 159ms/epoch - 4ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0666 - val_loss: 0.0387 - val_mse: 0.0387 - val_mae: 0.1422 - lr: 1.0000e-05 - 157ms/epoch - 4ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0669 - val_loss: 0.0387 - val_mse: 0.0387 - val_mae: 0.1423 - lr: 1.0000e-05 - 184ms/epoch - 4ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0669 - val_loss: 0.0387 - val_mse: 0.0387 - val_mae: 0.1423 - lr: 1.0000e-05 - 171ms/epoch - 4ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0685 - val_loss: 0.0387 - val_mse: 0.0387 - val_mae: 0.1424 - lr: 1.0000e-05 - 172ms/epoch - 4ms/step
Epoch 19/500
Epoch 00019: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00019: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0073 - mse: 0.0073 - mae: 0.0694 - val_loss: 0.0387 - val_mse: 0.0387 - val_mae: 0.1424 - lr: 1.0000e-05 - 171ms/epoch - 4ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0636 - val_loss: 0.0388 - val_mse: 0.0388 - val_mae: 0.1426 - lr: 1.0000e-05 - 162ms/epoch - 4ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0676 - val_loss: 0.0389 - val_mse: 0.0389 - val_mae: 0.1427 - lr: 1.0000e-05 - 169ms/epoch - 4ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0074 - mse: 0.0074 - mae: 0.0694 - val_loss: 0.0390 - val_mse: 0.0390 - val_mae: 0.1428 - lr: 1.0000e-05 - 173ms/epoch - 4ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0675 - val_loss: 0.0390 - val_mse: 0.0390 - val_mae: 0.1428 - lr: 1.0000e-05 - 165ms/epoch - 4ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0649 - val_loss: 0.0390 - val_mse: 0.0390 - val_mae: 0.1429 - lr: 1.0000e-05 - 178ms/epoch - 4ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0626 - val_loss: 0.0391 - val_mse: 0.0391 - val_mae: 0.1431 - lr: 1.0000e-05 - 169ms/epoch - 4ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0663 - val_loss: 0.0393 - val_mse: 0.0393 - val_mae: 0.1433 - lr: 1.0000e-05 - 164ms/epoch - 4ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0621 - val_loss: 0.0395 - val_mse: 0.0395 - val_mae: 0.1435 - lr: 1.0000e-05 - 180ms/epoch - 4ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0684 - val_loss: 0.0395 - val_mse: 0.0395 - val_mae: 0.1436 - lr: 1.0000e-05 - 181ms/epoch - 4ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0660 - val_loss: 0.0396 - val_mse: 0.0396 - val_mae: 0.1437 - lr: 1.0000e-05 - 168ms/epoch - 4ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0654 - val_loss: 0.0397 - val_mse: 0.0397 - val_mae: 0.1439 - lr: 1.0000e-05 - 166ms/epoch - 4ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0655 - val_loss: 0.0398 - val_mse: 0.0398 - val_mae: 0.1440 - lr: 1.0000e-05 - 164ms/epoch - 4ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0663 - val_loss: 0.0399 - val_mse: 0.0399 - val_mae: 0.1441 - lr: 1.0000e-05 - 162ms/epoch - 4ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0644 - val_loss: 0.0401 - val_mse: 0.0401 - val_mae: 0.1444 - lr: 1.0000e-05 - 181ms/epoch - 4ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0641 - val_loss: 0.0403 - val_mse: 0.0403 - val_mae: 0.1446 - lr: 1.0000e-05 - 167ms/epoch - 4ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0634 - val_loss: 0.0404 - val_mse: 0.0404 - val_mae: 0.1448 - lr: 1.0000e-05 - 166ms/epoch - 4ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0629 - val_loss: 0.0403 - val_mse: 0.0403 - val_mae: 0.1448 - lr: 1.0000e-05 - 167ms/epoch - 4ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0657 - val_loss: 0.0402 - val_mse: 0.0402 - val_mae: 0.1447 - lr: 1.0000e-05 - 174ms/epoch - 4ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0643 - val_loss: 0.0401 - val_mse: 0.0401 - val_mae: 0.1445 - lr: 1.0000e-05 - 162ms/epoch - 4ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0657 - val_loss: 0.0402 - val_mse: 0.0402 - val_mae: 0.1447 - lr: 1.0000e-05 - 182ms/epoch - 4ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0657 - val_loss: 0.0404 - val_mse: 0.0404 - val_mae: 0.1450 - lr: 1.0000e-05 - 176ms/epoch - 4ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0650 - val_loss: 0.0406 - val_mse: 0.0406 - val_mae: 0.1452 - lr: 1.0000e-05 - 171ms/epoch - 4ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0634 - val_loss: 0.0408 - val_mse: 0.0408 - val_mae: 0.1455 - lr: 1.0000e-05 - 167ms/epoch - 4ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0651 - val_loss: 0.0409 - val_mse: 0.0409 - val_mae: 0.1455 - lr: 1.0000e-05 - 163ms/epoch - 4ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0066 - mse: 0.0066 - mae: 0.0655 - val_loss: 0.0408 - val_mse: 0.0408 - val_mae: 0.1455 - lr: 1.0000e-05 - 170ms/epoch - 4ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0632 - val_loss: 0.0409 - val_mse: 0.0409 - val_mae: 0.1456 - lr: 1.0000e-05 - 168ms/epoch - 4ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0635 - val_loss: 0.0409 - val_mse: 0.0409 - val_mae: 0.1456 - lr: 1.0000e-05 - 166ms/epoch - 4ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0617 - val_loss: 0.0408 - val_mse: 0.0408 - val_mae: 0.1455 - lr: 1.0000e-05 - 160ms/epoch - 4ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0671 - val_loss: 0.0408 - val_mse: 0.0408 - val_mae: 0.1455 - lr: 1.0000e-05 - 163ms/epoch - 4ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0616 - val_loss: 0.0410 - val_mse: 0.0410 - val_mae: 0.1458 - lr: 1.0000e-05 - 157ms/epoch - 4ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0664 - val_loss: 0.0410 - val_mse: 0.0410 - val_mae: 0.1457 - lr: 1.0000e-05 - 171ms/epoch - 4ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0628 - val_loss: 0.0410 - val_mse: 0.0410 - val_mae: 0.1458 - lr: 1.0000e-05 - 189ms/epoch - 4ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0634 - val_loss: 0.0412 - val_mse: 0.0412 - val_mae: 0.1461 - lr: 1.0000e-05 - 169ms/epoch - 4ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0646 - val_loss: 0.0414 - val_mse: 0.0414 - val_mae: 0.1463 - lr: 1.0000e-05 - 164ms/epoch - 4ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0642 - val_loss: 0.0413 - val_mse: 0.0413 - val_mae: 0.1462 - lr: 1.0000e-05 - 158ms/epoch - 4ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0635 - val_loss: 0.0414 - val_mse: 0.0414 - val_mae: 0.1463 - lr: 1.0000e-05 - 166ms/epoch - 4ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0615 - val_loss: 0.0416 - val_mse: 0.0416 - val_mae: 0.1466 - lr: 1.0000e-05 - 162ms/epoch - 4ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0615 - val_loss: 0.0417 - val_mse: 0.0417 - val_mae: 0.1468 - lr: 1.0000e-05 - 171ms/epoch - 4ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0605 - val_loss: 0.0420 - val_mse: 0.0420 - val_mae: 0.1472 - lr: 1.0000e-05 - 167ms/epoch - 4ms/step
Epoch 59/500
Epoch 00059: val_loss did not improve from 0.03334
43/43 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0634 - val_loss: 0.0423 - val_mse: 0.0423 - val_mae: 0.1476 - lr: 1.0000e-05 - 171ms/epoch - 4ms/step
Epoch 00059: early stopping
SMA
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 37.148636819682245
RMSE: 6.094968155756209
MAPE: 5.090179331518223
EMA
Prediction vs Close: 55.22% Accuracy
Prediction vs Prediction: 47.39% Accuracy
MSE: 31.582654654902484
RMSE: 5.619844718041815
MAPE: 4.507182634072088
WMA
Prediction vs Close: 52.99% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 65.00101872981564
RMSE: 8.062320926992156
MAPE: 6.705711592581163
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.52% Accuracy
MSE: 35.269002386685244
RMSE: 5.938771117553298
MAPE: 4.62878838931535
KAMA
Prediction vs Close: 54.1% Accuracy
Prediction vs Prediction: 48.88% Accuracy
MSE: 62.25156682504816
RMSE: 7.8899662119078915
MAPE: 6.222956717810362
MIDPOINT
Prediction vs Close: 51.12% Accuracy
Prediction vs Prediction: 42.54% Accuracy
MSE: 84.47531990876952
RMSE: 9.191045637399997
MAPE: 7.890202393641488
T3
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 92.6070950307407
RMSE: 9.623258025780078
MAPE: 8.212466968891306
TEMA
TEMA([input_arrays], [timeperiod=30])
Triple Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
9
Working on TEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-16493.570, Time=2.59 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-15527.581, Time=7.51 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16154.477, Time=7.44 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-15134.948, Time=6.94 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-16538.454, Time=8.24 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-16271.346, Time=2.18 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-16350.992, Time=13.28 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-16200.149, Time=3.30 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-16461.809, Time=15.38 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-16384.147, Time=3.00 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=10.18 sec
ARIMA(2,3,1)(0,0,0)[0] intercept : AIC=-15110.164, Time=5.74 sec
Best model: ARIMA(2,3,1)(0,0,0)[0]
Total fit time: 85.805 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(2, 3, 1) Log Likelihood 8294.227
Date: Sun, 12 Dec 2021 AIC -16538.454
Time: 15:40:52 BIC -16421.183
Sample: 0 HQIC -16493.417
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 3.591e-07 0.001 0.000 1.000 -0.002 0.002
x2 3.6e-07 0.002 0.000 1.000 -0.003 0.003
x3 3.611e-07 0.001 0.000 1.000 -0.002 0.002
x4 1.0000 0.000 2628.605 0.000 0.999 1.001
x5 3.432e-07 0.000 0.001 0.999 -0.001 0.001
x6 1.714e-07 4.05e-05 0.004 0.997 -7.91e-05 7.95e-05
x7 3.541e-07 0.001 0.000 1.000 -0.003 0.003
x8 -0.0002 0.000 -1.006 0.315 -0.001 0.000
x9 -7.559e-08 0.000 -0.000 1.000 -0.001 0.001
x10 0.0001 0.000 0.492 0.623 -0.000 0.001
x11 -0.0006 0.000 -2.697 0.007 -0.001 -0.000
x12 0.0005 0.000 1.741 0.082 -5.97e-05 0.001
x13 3.6e-07 0.000 0.002 0.999 -0.000 0.000
x14 1.003e-06 0.001 0.001 0.999 -0.002 0.002
x15 3.506e-07 7.16e-05 0.005 0.996 -0.000 0.000
x16 5.157e-07 0.000 0.005 0.996 -0.000 0.000
x17 3.516e-07 6.59e-05 0.005 0.996 -0.000 0.000
x18 1.166e-07 0.000 0.001 1.000 -0.000 0.000
x19 3.922e-07 7.5e-05 0.005 0.996 -0.000 0.000
x20 -3.64e-07 0.000 -0.002 0.999 -0.000 0.000
x21 4.458e-07 0.000 0.004 0.997 -0.000 0.000
ar.L1 -0.4019 4.12e-05 -9758.484 0.000 -0.402 -0.402
ar.L2 -0.1006 1.58e-05 -6360.873 0.000 -0.101 -0.101
ma.L1 -0.7963 8.45e-06 -9.43e+04 0.000 -0.796 -0.796
sigma2 9.048e-11 7.2e-11 1.257 0.209 -5.06e-11 2.32e-10
===================================================================================
Ljung-Box (L1) (Q): 64.02 Jarque-Bera (JB): 4424775.47
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.53
Prob(H) (two-sided): 0.00 Kurtosis: 366.04
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.02e+20. Standard errors may be unstable.
ARIMA order: (2, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.06351, saving model to LSTM7.h5
90/90 - 2s - loss: 0.1164 - mse: 0.1164 - mae: 0.2369 - val_loss: 0.0635 - val_mse: 0.0635 - val_mae: 0.2342 - lr: 0.0010 - 2s/epoch - 26ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.06351
90/90 - 0s - loss: 0.0217 - mse: 0.0217 - mae: 0.1164 - val_loss: 0.0752 - val_mse: 0.0752 - val_mae: 0.2599 - lr: 0.0010 - 322ms/epoch - 4ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.06351 to 0.03991, saving model to LSTM7.h5
90/90 - 0s - loss: 0.0146 - mse: 0.0146 - mae: 0.0949 - val_loss: 0.0399 - val_mse: 0.0399 - val_mae: 0.1852 - lr: 0.0010 - 333ms/epoch - 4ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.03991 to 0.02556, saving model to LSTM7.h5
90/90 - 0s - loss: 0.0109 - mse: 0.0109 - mae: 0.0820 - val_loss: 0.0256 - val_mse: 0.0256 - val_mae: 0.1452 - lr: 0.0010 - 334ms/epoch - 4ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.02556
90/90 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0674 - val_loss: 0.0340 - val_mse: 0.0340 - val_mae: 0.1718 - lr: 0.0010 - 321ms/epoch - 4ms/step
Epoch 6/500
Epoch 00006: val_loss improved from 0.02556 to 0.01074, saving model to LSTM7.h5
90/90 - 0s - loss: 0.0091 - mse: 0.0091 - mae: 0.0717 - val_loss: 0.0107 - val_mse: 0.0107 - val_mae: 0.0886 - lr: 0.0010 - 322ms/epoch - 4ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.01074
90/90 - 0s - loss: 0.0113 - mse: 0.0113 - mae: 0.0827 - val_loss: 0.0345 - val_mse: 0.0345 - val_mae: 0.1738 - lr: 0.0010 - 324ms/epoch - 4ms/step
Epoch 8/500
Epoch 00008: val_loss improved from 0.01074 to 0.00726, saving model to LSTM7.h5
90/90 - 0s - loss: 0.0197 - mse: 0.0197 - mae: 0.1079 - val_loss: 0.0073 - val_mse: 0.0073 - val_mae: 0.0705 - lr: 0.0010 - 330ms/epoch - 4ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0265 - mse: 0.0265 - mae: 0.1264 - val_loss: 0.0524 - val_mse: 0.0524 - val_mae: 0.2111 - lr: 0.0010 - 324ms/epoch - 4ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0165 - mse: 0.0165 - mae: 0.0981 - val_loss: 0.0132 - val_mse: 0.0132 - val_mae: 0.0965 - lr: 0.0010 - 330ms/epoch - 4ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0090 - mse: 0.0090 - mae: 0.0742 - val_loss: 0.0873 - val_mse: 0.0873 - val_mae: 0.2716 - lr: 0.0010 - 313ms/epoch - 3ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0619 - val_loss: 0.0194 - val_mse: 0.0194 - val_mae: 0.1182 - lr: 0.0010 - 323ms/epoch - 4ms/step
Epoch 13/500
Epoch 00013: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00013: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0618 - val_loss: 0.1000 - val_mse: 0.1000 - val_mae: 0.2899 - lr: 0.0010 - 314ms/epoch - 3ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0102 - mse: 0.0102 - mae: 0.0809 - val_loss: 0.0546 - val_mse: 0.0546 - val_mae: 0.2069 - lr: 1.0000e-04 - 306ms/epoch - 3ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0528 - val_loss: 0.0485 - val_mse: 0.0485 - val_mae: 0.1932 - lr: 1.0000e-04 - 325ms/epoch - 4ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0543 - val_loss: 0.0455 - val_mse: 0.0455 - val_mae: 0.1859 - lr: 1.0000e-04 - 313ms/epoch - 3ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0513 - val_loss: 0.0446 - val_mse: 0.0446 - val_mae: 0.1836 - lr: 1.0000e-04 - 327ms/epoch - 4ms/step
Epoch 18/500
Epoch 00018: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00018: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0539 - val_loss: 0.0432 - val_mse: 0.0432 - val_mae: 0.1800 - lr: 1.0000e-04 - 310ms/epoch - 3ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0478 - val_loss: 0.0425 - val_mse: 0.0425 - val_mae: 0.1782 - lr: 1.0000e-05 - 321ms/epoch - 4ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0512 - val_loss: 0.0417 - val_mse: 0.0417 - val_mae: 0.1763 - lr: 1.0000e-05 - 331ms/epoch - 4ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0503 - val_loss: 0.0416 - val_mse: 0.0416 - val_mae: 0.1760 - lr: 1.0000e-05 - 310ms/epoch - 3ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0501 - val_loss: 0.0408 - val_mse: 0.0408 - val_mae: 0.1740 - lr: 1.0000e-05 - 313ms/epoch - 3ms/step
Epoch 23/500
Epoch 00023: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00023: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0503 - val_loss: 0.0405 - val_mse: 0.0405 - val_mae: 0.1734 - lr: 1.0000e-05 - 315ms/epoch - 3ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0489 - val_loss: 0.0403 - val_mse: 0.0403 - val_mae: 0.1727 - lr: 1.0000e-05 - 329ms/epoch - 4ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0473 - val_loss: 0.0400 - val_mse: 0.0400 - val_mae: 0.1720 - lr: 1.0000e-05 - 315ms/epoch - 4ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0522 - val_loss: 0.0400 - val_mse: 0.0400 - val_mae: 0.1719 - lr: 1.0000e-05 - 310ms/epoch - 3ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0500 - val_loss: 0.0397 - val_mse: 0.0397 - val_mae: 0.1712 - lr: 1.0000e-05 - 312ms/epoch - 3ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0490 - val_loss: 0.0398 - val_mse: 0.0398 - val_mae: 0.1713 - lr: 1.0000e-05 - 309ms/epoch - 3ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0488 - val_loss: 0.0398 - val_mse: 0.0398 - val_mae: 0.1713 - lr: 1.0000e-05 - 317ms/epoch - 4ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0475 - val_loss: 0.0402 - val_mse: 0.0402 - val_mae: 0.1723 - lr: 1.0000e-05 - 315ms/epoch - 3ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0502 - val_loss: 0.0401 - val_mse: 0.0401 - val_mae: 0.1721 - lr: 1.0000e-05 - 320ms/epoch - 4ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0501 - val_loss: 0.0405 - val_mse: 0.0405 - val_mae: 0.1729 - lr: 1.0000e-05 - 333ms/epoch - 4ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0489 - val_loss: 0.0407 - val_mse: 0.0407 - val_mae: 0.1735 - lr: 1.0000e-05 - 317ms/epoch - 4ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0484 - val_loss: 0.0409 - val_mse: 0.0409 - val_mae: 0.1739 - lr: 1.0000e-05 - 319ms/epoch - 4ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0488 - val_loss: 0.0406 - val_mse: 0.0406 - val_mae: 0.1730 - lr: 1.0000e-05 - 320ms/epoch - 4ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0483 - val_loss: 0.0407 - val_mse: 0.0407 - val_mae: 0.1732 - lr: 1.0000e-05 - 313ms/epoch - 3ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0509 - val_loss: 0.0409 - val_mse: 0.0409 - val_mae: 0.1737 - lr: 1.0000e-05 - 307ms/epoch - 3ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0508 - val_loss: 0.0413 - val_mse: 0.0413 - val_mae: 0.1747 - lr: 1.0000e-05 - 316ms/epoch - 4ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0479 - val_loss: 0.0414 - val_mse: 0.0414 - val_mae: 0.1749 - lr: 1.0000e-05 - 323ms/epoch - 4ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0485 - val_loss: 0.0410 - val_mse: 0.0410 - val_mae: 0.1738 - lr: 1.0000e-05 - 312ms/epoch - 3ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0482 - val_loss: 0.0413 - val_mse: 0.0413 - val_mae: 0.1745 - lr: 1.0000e-05 - 307ms/epoch - 3ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0497 - val_loss: 0.0415 - val_mse: 0.0415 - val_mae: 0.1750 - lr: 1.0000e-05 - 320ms/epoch - 4ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0478 - val_loss: 0.0415 - val_mse: 0.0415 - val_mae: 0.1747 - lr: 1.0000e-05 - 309ms/epoch - 3ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0480 - val_loss: 0.0414 - val_mse: 0.0414 - val_mae: 0.1745 - lr: 1.0000e-05 - 321ms/epoch - 4ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0487 - val_loss: 0.0413 - val_mse: 0.0413 - val_mae: 0.1743 - lr: 1.0000e-05 - 322ms/epoch - 4ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0507 - val_loss: 0.0408 - val_mse: 0.0408 - val_mae: 0.1730 - lr: 1.0000e-05 - 315ms/epoch - 3ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0478 - val_loss: 0.0409 - val_mse: 0.0409 - val_mae: 0.1733 - lr: 1.0000e-05 - 306ms/epoch - 3ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0480 - val_loss: 0.0405 - val_mse: 0.0405 - val_mae: 0.1723 - lr: 1.0000e-05 - 319ms/epoch - 4ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0492 - val_loss: 0.0412 - val_mse: 0.0412 - val_mae: 0.1738 - lr: 1.0000e-05 - 323ms/epoch - 4ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0504 - val_loss: 0.0416 - val_mse: 0.0416 - val_mae: 0.1747 - lr: 1.0000e-05 - 325ms/epoch - 4ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0041 - mse: 0.0041 - mae: 0.0499 - val_loss: 0.0421 - val_mse: 0.0421 - val_mae: 0.1758 - lr: 1.0000e-05 - 311ms/epoch - 3ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0482 - val_loss: 0.0426 - val_mse: 0.0426 - val_mae: 0.1770 - lr: 1.0000e-05 - 318ms/epoch - 4ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0038 - mse: 0.0038 - mae: 0.0484 - val_loss: 0.0432 - val_mse: 0.0432 - val_mae: 0.1784 - lr: 1.0000e-05 - 330ms/epoch - 4ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0036 - mse: 0.0036 - mae: 0.0465 - val_loss: 0.0432 - val_mse: 0.0432 - val_mae: 0.1782 - lr: 1.0000e-05 - 330ms/epoch - 4ms/step
Epoch 55/500
Epoch 00055: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0040 - mse: 0.0040 - mae: 0.0494 - val_loss: 0.0426 - val_mse: 0.0426 - val_mae: 0.1768 - lr: 1.0000e-05 - 310ms/epoch - 3ms/step
Epoch 56/500
Epoch 00056: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0485 - val_loss: 0.0432 - val_mse: 0.0432 - val_mae: 0.1782 - lr: 1.0000e-05 - 314ms/epoch - 3ms/step
Epoch 57/500
Epoch 00057: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0037 - mse: 0.0037 - mae: 0.0473 - val_loss: 0.0426 - val_mse: 0.0426 - val_mae: 0.1766 - lr: 1.0000e-05 - 318ms/epoch - 4ms/step
Epoch 58/500
Epoch 00058: val_loss did not improve from 0.00726
90/90 - 0s - loss: 0.0039 - mse: 0.0039 - mae: 0.0493 - val_loss: 0.0435 - val_mse: 0.0435 - val_mae: 0.1787 - lr: 1.0000e-05 - 340ms/epoch - 4ms/step
Epoch 00058: early stopping
SMA Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 48.88% Accuracy MSE: 37.148636819682245 RMSE: 6.094968155756209 MAPE: 5.090179331518223 EMA Prediction vs Close: 55.22% Accuracy Prediction vs Prediction: 47.39% Accuracy MSE: 31.582654654902484 RMSE: 5.619844718041815 MAPE: 4.507182634072088 WMA Prediction vs Close: 52.99% Accuracy Prediction vs Prediction: 48.88% Accuracy MSE: 65.00101872981564 RMSE: 8.062320926992156 MAPE: 6.705711592581163 DEMA Prediction vs Close: 52.24% Accuracy Prediction vs Prediction: 45.52% Accuracy MSE: 35.269002386685244 RMSE: 5.938771117553298 MAPE: 4.62878838931535 KAMA Prediction vs Close: 54.1% Accuracy Prediction vs Prediction: 48.88% Accuracy MSE: 62.25156682504816 RMSE: 7.8899662119078915 MAPE: 6.222956717810362 MIDPOINT Prediction vs Close: 51.12% Accuracy Prediction vs Prediction: 42.54% Accuracy MSE: 84.47531990876952 RMSE: 9.191045637399997 MAPE: 7.890202393641488 T3 Prediction vs Close: 54.85% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 92.6070950307407 RMSE: 9.623258025780078 MAPE: 8.212466968891306 TEMA Prediction vs Close: 51.49% Accuracy Prediction vs Prediction: 47.01% Accuracy MSE: 50.229608880159034 RMSE: 7.087285014740061 MAPE: 6.2877236827531835 Runtime: mins: 42.43883018304999
Architecture Used¶
In [ ]:
from google.colab import files
import cv2
uploaded = files.upload()
In [ ]:
img = cv2.imread('Experiment5.png')
plt.figure(figsize=(20,10))
plt.axis("off")
plt.title('LSTM Architecture '+imgfile,fontsize=18)
plt.imshow(img)
Out[ ]:
<matplotlib.image.AxesImage at 0x7fda6d2f7ed0>
Model Plots¶
In [83]:
with open('simulation7_data.json') as json_file:
simulation7 = json.load(json_file)
fileimg = 'Experiment7'
In [84]:
for i in range(len(list(simulation7.keys()))):
SIM = list(simulation7.keys())[i]
plot_train(simulation7,SIM)
plot_test(simulation7,SIM)
----- Train RMSE for SMA ----- 8.955057105534777 ----- Train_MSE_LSTM for SMA ----- 80.19304776338892 ----- Train MAE LSTM for SMA ----- 7.794913717002681
----- Test RMSE for SMA----- 6.094968155756209 ----- Test_MSE_LSTM for SMA----- 37.148636819682245 ----- Test_MAE_LSTM for SMA----- 5.090179331518223
----- Train RMSE for EMA ----- 10.61356917187827 ----- Train_MSE_LSTM for EMA ----- 112.6478505662448 ----- Train MAE LSTM for EMA ----- 9.447697131450768
----- Test RMSE for EMA----- 5.619844718041815 ----- Test_MSE_LSTM for EMA----- 31.582654654902484 ----- Test_MAE_LSTM for EMA----- 4.507182634072088
----- Train RMSE for WMA ----- 10.751551158526922 ----- Train_MSE_LSTM for WMA ----- 115.59585231442159 ----- Train MAE LSTM for WMA ----- 9.750787083557483
----- Test RMSE for WMA----- 8.062320926992156 ----- Test_MSE_LSTM for WMA----- 65.00101872981564 ----- Test_MAE_LSTM for WMA----- 6.705711592581163
----- Train RMSE for DEMA ----- 12.583936483896219 ----- Train_MSE_LSTM for DEMA ----- 158.3554574307343 ----- Train MAE LSTM for DEMA ----- 11.414769962937156
----- Test RMSE for DEMA----- 5.938771117553298 ----- Test_MSE_LSTM for DEMA----- 35.269002386685244 ----- Test_MAE_LSTM for DEMA----- 4.62878838931535
----- Train RMSE for KAMA ----- 10.911962583126925 ----- Train_MSE_LSTM for KAMA ----- 119.07092741556205 ----- Train MAE LSTM for KAMA ----- 9.89734832853964
----- Test RMSE for KAMA----- 7.8899662119078915 ----- Test_MSE_LSTM for KAMA----- 62.25156682504816 ----- Test_MAE_LSTM for KAMA----- 6.222956717810362
----- Train RMSE for MIDPOINT ----- 9.704593247771532 ----- Train_MSE_LSTM for MIDPOINT ----- 94.17913010469283 ----- Train MAE LSTM for MIDPOINT ----- 8.665870629654135
----- Test RMSE for MIDPOINT----- 9.191045637399997 ----- Test_MSE_LSTM for MIDPOINT----- 84.47531990876952 ----- Test_MAE_LSTM for MIDPOINT----- 7.890202393641488
----- Train RMSE for T3 ----- 12.326458508446358 ----- Train_MSE_LSTM for T3 ----- 151.94157936044962 ----- Train MAE LSTM for T3 ----- 11.126900240195694
----- Test RMSE for T3----- 9.623258025780078 ----- Test_MSE_LSTM for T3----- 92.6070950307407 ----- Test_MAE_LSTM for T3----- 8.212466968891306
----- Train RMSE for TEMA ----- 7.3831693218457195 ----- Train_MSE_LSTM for TEMA ----- 54.51118923504378 ----- Train MAE LSTM for TEMA ----- 5.0852937426664155
----- Test RMSE for TEMA----- 7.087285014740061 ----- Test_MSE_LSTM for TEMA----- 50.229608880159034 ----- Test_MAE_LSTM for TEMA----- 6.2877236827531835
Arima w Exogenous Variable Multistep MutiVariate LSTM Hybrid Model Experiment 8¶
In [ ]:
def get_arima_exog(dataframe,original_data, train_len, test_len):
# prepare train and test data for exogenous vr
X_value = pd.DataFrame(low_vol.iloc[:, :])
y_value = pd.DataFrame(low_vol.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
X_scale_dataset = X_scaler.fit_transform(X_value)
y_scale_dataset = y_scaler.fit_transform(y_value)
# Get data and check shape
# X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X_scale_dataset)
y_train, y_test, = split_train_test(y_scale_dataset)
yc_train,yc_test = split_train_test(low_vol_data)
yc = yc_test.values.tolist()
y_train_list = y_train.flatten().tolist()
y_test_list = y_test.flatten().tolist()
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
# Initialize model
model = auto_arima(y_train_list,exogenous = X_train,trace=True, error_action='ignore', start_p=1,start_q=1,max_p=3,max_q=3,d=3,
suppress_warnings=True,stepwise=True,seasonal=True)
# Determine model parameters
print(model.summary())
model.fit(y_train_list,maxiter=200)
order = model.get_params()['order']
print('ARIMA order:', order, '\n')
# Genereate predictions
prediction = []
for i in range(len(y_test_list)):
model = pmdarima.ARIMA(order=order)
model.fit(y_train_list)
# print('working on', i+1, 'of', len(y_test), '-- ' + str(int(100 * (i + 1) / len(y_test))) + '% complete')
prediction.append(model.predict()[0])
y_train_list.append(y_test_list[i])
predictionte = y_scaler.inverse_transform(np.array(prediction).reshape(-1,1))
y_test_ = y_scaler.inverse_transform(np.array(y_test_list).reshape(-1,1))
# Generate error data
mse = mean_squared_error(yc_test, predictionte)
rmse = mse ** 0.5
mae = mean_absolute_error(y_test_ , predictionte )
return yc,predictionte.flatten().tolist(), mse, rmse, mae
In [ ]:
def get_lstm(data,original_data, train_len, test_len,img_file,ma ,lstm_len=3):
# prepare train and test data
X_value = pd.DataFrame(data.iloc[:, :])
y_value = pd.DataFrame(data.iloc[:, 3])
X_scaler = MinMaxScaler(feature_range=(-1, 1))
y_scaler = MinMaxScaler(feature_range=(-1, 1))
X_scaler.fit(X_value)
y_scaler.fit(y_value)
# Get data and check shape
X, y, yc = get_X_y(X_scale_dataset, y_scale_dataset)#X will be of shape 224 X 3 X 21 (each 3 X 21 array will be 3 days' worth of data). yc will have the corresponding closing price value
# pdb.set_trace()
X_train, X_test, = split_train_test(X)
y_train, y_test, = split_train_test(y)
# yc_train, yc_test, = split_train_test(original_data)
index_train, index_test, = predict_index(dataset_final, X_train, n_steps_in, n_steps_out)
det =20
input_dim = X_train.shape[1]#3
feature_size = X_train.shape[2]#24
output_dim = y_train.shape[1]#1
# Option 1
# Set up & fit LSTM RNN
# model = Sequential()
# model.add(LSTM(256, activation='relu', kernel_initializer='he_normal', input_shape=(input_dim, feature_size)))
# model.add(Dense(units=64,activation='relu'))
# model.add(Dropout(0.5))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(learning_rate = 0.001), loss='mse')
# ## Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM1.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma_' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# # # option 2
# model = Sequential()
# model.add(Bidirectional(LSTM(units= 128), input_shape=(input_dim, feature_size)))
# model.add(Dense(64))
# model.add(Dense(units=output_dim))
# model.compile(optimizer=Adam(lr = 0.001), loss='mean_squared_error', metrics=['accuracy'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM7.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma+' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# Option 3
# define custom activation
#
# class Double_Tanh(Activation):
# def __init__(self, activation, **kwargs):
# super(Double_Tanh, self).__init__(activation, **kwargs)
# self.__name__ = 'double_tanh'
# def double_tanh(x):
# return (K.tanh(x) * 2)
# get_custom_objects().update({'double_tanh':Double_Tanh(double_tanh)})
# # Model Generation
# model = Sequential()
# #check https://machinelearningmastery.com/use-weight-regularization-lstm-networks-time-series-forecasting/
# model.add(LSTM(25, input_shape=(input_dim, feature_size), dropout=0.2, kernel_regularizer=l1_l2(0.00,0.00), bias_regularizer=l1_l2(0.00,0.00)))
# model.add(Dense(1))
# model.add(Activation(double_tanh))
# model.compile(loss='mean_squared_error', optimizer='adam', metrics=['mse', 'mae'])
# # Common code
# callbacks = [
# EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
# ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
# ModelCheckpoint('LSTM7.h5', verbose=1, save_best_only=True, save_weights_only=True)]
# fname1 = img_file+'.png'
# tensorflow.keras.utils.plot_model(
# model, to_file=fname1, show_shapes=True, show_dtype=False,
# show_layer_names=True, expand_nested=False, dpi=96,
# layer_range=None, show_layer_activations=False
# )
# history = model.fit(X_train, y_train, epochs=500, batch_size=1, verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# # plot loss
# fname2 = img_file+'-'+ma
# plt.title(img_file+'-'+ma+' Loss')
# plt.xlabel("Epochs")
# plt.ylabel("Loss")
# pyplot.plot(history.history['loss'], label='train')
# pyplot.plot(history.history['val_loss'], label='validation')
# pyplot.legend()
# pyplot.savefig(fname2+'.png',dpi='figure')
# pyplot.show()
# #Option 4
# # Set up & fit LSTM RNN
model = Sequential()
model.add(LSTM(units=lstm_len, return_sequences=True, input_shape=(input_dim, feature_size)))
model.add(LSTM(units=int(lstm_len/2)))
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='mean_squared_error', optimizer='adam')
# Common code
callbacks = [
EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=50),
ReduceLROnPlateau(factor=0.1, patience=5, min_lr=0.00001, verbose=1),
ModelCheckpoint('LSTM8.h5', verbose=1, save_best_only=True, save_weights_only=True)]
fname1 = img_file+'.png'
tensorflow.keras.utils.plot_model(
model, to_file=fname1, show_shapes=True, show_dtype=False,
show_layer_names=True, expand_nested=False, dpi=96,
layer_range=None, show_layer_activations=False
)
history = model.fit(X_train, y_train, epochs=500, batch_size=int( optimized_period[ma]), verbose=2, callbacks=callbacks, validation_data=(X_test, y_test),shuffle=False)
# plot loss
fname2 = img_file+'-'+ma
plt.title(img_file+'-'+ma+' Loss')
plt.xlabel("Epochs")
plt.ylabel("Loss")
pyplot.plot(history.history['loss'], label='train')
pyplot.plot(history.history['val_loss'], label='validation')
pyplot.legend()
pyplot.savefig(fname2+'.png',dpi='figure')
pyplot.show()
# Generate predictions
predictiontr = model.predict(X_train, verbose=0)
predictiontr = y_scaler.inverse_transform(predictiontr).tolist()
outputtr = []
for i in range(len(predictiontr)):
outputtr.extend(predictiontr[i])
predictiontr = outputtr
# Generate error data
## replace with yc , xtest generated by new multistep method
mse_tr = mean_squared_error(y_train, predictiontr)
rmse_tr = mse_tr ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictiontr))
mae_tr = mean_absolute_error(y_train, pd.Series(predictiontr))
# Original_tr = pd.Series(yc_train)
Original_tr = y_scaler.inverse_transform(y_train).flatten().tolist()
predictionte = model.predict(X_test, verbose=0)
predictionte = (y_scaler.inverse_transform(predictionte)-det).tolist()
outputte = []
for i in range(len(predictionte)):
outputte.extend(predictionte[i])
predictionte = outputte
# Generate error data
mse_te = mean_squared_error(y_test, predictionte)
rmse_te = mse_te ** 0.5
# mape = mean_absolute_percentage_error(X_test, pd.Series(predictionte))
mae_te = mean_absolute_error(y_test, pd.Series(predictionte))
# Original_te = pd.Series(yc_test)
Original_te = y_scaler.inverse_transform(y_test).flatten().tolist()
return Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,Original_te,predictionte, mse_te,rmse_te,mae_te
In [ ]:
if __name__ == '__main__':
start_time = timeit.default_timer()
simulation8 = {}
imgfile = 'Experiment8'
for ma in optimized_period:
print(ma)
print(functions[ma])
print ( int( optimized_period[ma]))
# if ma == 'SMA':
low_vol = df.apply(lambda c: functions[ma](c, timeperiod = int( optimized_period[ma])))
low_vol = low_vol.fillna(0)
low_vol_data = df['close']
high_vol = pd.DataFrame()
df2 = df.copy()
for i in df2.columns:
if i in low_vol.columns:
high_vol[i] = df2[i].subtract(low_vol[i], fill_value=0)
high_vol_data = df['close']
## *****************************************************
# Generate ARIMA and LSTM predictions
print('\nWorking on ' + ma + ' predictions')
try:
print('parameters used : ', train_len, test_len)
low_vol_Original, low_vol_prediction, low_vol_mse, low_vol_rmse,low_vol_mae = get_arima_exog(low_vol,low_vol_data, train_len, test_len)
except:
print('ARIMA error, skipping to next MA type')
continue
Original_tr, predictiontr, mse_tr, rmse_tr,mae_tr,high_vol_Original, high_vol_prediction, high_vol_mse, high_vol_rmse,high_vol_mae, = get_lstm(high_vol,high_vol_data, train_len, test_len,imgfile,ma)
final_prediction_tr = df['close'].head(train_len).values + pd.Series(predictiontr) # ignoring first 3 steps
mse_ftr = mean_squared_error(df['close'].head(train_len).values,final_prediction_tr.values)
rmse_ftr = mse_ftr ** 0.5
mape_ftr = mean_absolute_percentage_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
mae_ftr = mean_absolute_error(df['close'].head(train_len).reset_index(drop=True), final_prediction_tr)
final_prediction = pd.Series(low_vol_prediction[3:]) + pd.Series(high_vol_prediction)
mse = mean_squared_error(df['close'].tail(test_len).values,final_prediction.values)
rmse = mse ** 0.5
mape = mean_absolute_percentage_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
mae = mean_absolute_error(df['close'].tail(test_len).reset_index(drop=True), final_prediction)
# Generate prediction accuracy
actual = df['close'].tail(test_len).values
result_1 = []
result_2 = []
for i in range(1, len(final_prediction)):
# Compare prediction to previous close price
if final_prediction[i] > actual[i-1] and actual[i] > actual[i-1]:
result_1.append(1)
elif final_prediction[i] < actual[i-1] and actual[i] < actual[i-1]:
result_1.append(1)
else:
result_1.append(0)
# Compare prediction to previous prediction
if final_prediction[i] > final_prediction[i-1] and actual[i] > actual[i-1]:
result_2.append(1)
elif final_prediction[i] < final_prediction[i-1] and actual[i] < actual[i-1]:
result_2.append(1)
else:
result_2.append(0)
accuracy_1 = np.mean(result_1)
accuracy_2 = np.mean(result_2)
simulation8[ma] = {'low_vol': {'original':list(low_vol_Original), 'prediction': list(low_vol_prediction) , 'mse': low_vol_mse,
'rmse': low_vol_rmse, 'mae' : low_vol_mae},
'high_vol': {'original':list(high_vol_Original),'prediction': list(high_vol_prediction), 'mse': high_vol_mse,
'rmse': high_vol_rmse, 'mae' : high_vol_mae},
'final_tr': {'original':df['close'].head(train_len).tolist(),'prediction': final_prediction_tr.values.tolist(), 'mse': mse_ftr,
'rmse': rmse_ftr, 'mae' : mae_ftr},
'final': {'original': df['close'].tail(test_len).tolist(), 'prediction': final_prediction.values.tolist(), 'mse': mse,
'rmse': rmse, 'mae': mae },
'accuracy': {'prediction vs close': accuracy_1, 'prediction vs prediction': accuracy_2}}
# save simulation data here as checkpoint
with open('simulation8_data.json', 'w') as fp:
json.dump(simulation8, fp)
for ma in simulation8.keys():
print('\n' + ma)
print('Prediction vs Close:\t\t' + str(round(100*simulation8[ma]['accuracy']['prediction vs close'], 2))
+ '% Accuracy')
print('Prediction vs Prediction:\t' + str(round(100*simulation8[ma]['accuracy']['prediction vs prediction'], 2))
+ '% Accuracy')
print('MSE:\t', simulation8[ma]['final']['mse'],
'\nRMSE:\t', simulation8[ma]['final']['rmse'],
'\nMAPE:\t', simulation8[ma]['final']['mae'])#,
# '\nMAPE:\t', simulation[ma]['final']['mape'])
# else:
# break
elapsed = timeit.default_timer() - start_time
print('Runtime: mins:',elapsed/60)
SMA
SMA([input_arrays], [timeperiod=30])
Simple Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
17
Working on SMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-15000.708, Time=9.21 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-13492.284, Time=2.24 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15827.971, Time=8.03 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-13635.197, Time=10.27 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-14132.778, Time=3.58 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15140.312, Time=9.97 sec
ARIMA(1,3,0)(0,0,0)[0] intercept : AIC=-13970.469, Time=7.12 sec
Best model: ARIMA(1,3,0)(0,0,0)[0]
Total fit time: 50.431 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(1, 3, 0) Log Likelihood 7936.985
Date: Sun, 12 Dec 2021 AIC -15827.971
Time: 15:53:06 BIC -15720.081
Sample: 0 HQIC -15786.537
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -4.786e-05 0.001 -0.066 0.947 -0.001 0.001
x2 -4.789e-05 0.001 -0.085 0.932 -0.001 0.001
x3 -4.819e-05 0.000 -0.105 0.917 -0.001 0.001
x4 1.0000 0.001 1557.248 0.000 0.999 1.001
x5 -4.579e-05 0.001 -0.071 0.943 -0.001 0.001
x6 -5.16e-05 0.000 -0.432 0.666 -0.000 0.000
x7 -4.778e-05 0.000 -0.278 0.781 -0.000 0.000
x8 -0.0012 0.000 -7.403 0.000 -0.002 -0.001
x9 -3.454e-06 0.002 -0.002 0.998 -0.003 0.003
x10 -0.0005 0.001 -0.403 0.687 -0.003 0.002
x11 0.0029 0.000 10.904 0.000 0.002 0.003
x12 -0.0003 0.000 -1.815 0.069 -0.001 2.06e-05
x13 -4.809e-05 0.000 -0.157 0.875 -0.001 0.001
x14 -0.0001 0.000 -0.482 0.630 -0.001 0.000
x15 -5.214e-05 0.000 -0.273 0.785 -0.000 0.000
x16 -4.468e-05 0.000 -0.125 0.901 -0.001 0.001
x17 -4.224e-05 0.000 -0.202 0.840 -0.000 0.000
x18 -8.086e-05 0.000 -0.270 0.787 -0.001 0.001
x19 -5.537e-05 0.000 -0.244 0.807 -0.000 0.000
x20 8.423e-05 0.000 0.333 0.739 -0.000 0.001
x21 -4.232e-05 0.000 -0.166 0.868 -0.001 0.000
ar.L1 -0.6666 6.03e-06 -1.11e+05 0.000 -0.667 -0.667
sigma2 4.093e-10 8.97e-11 4.563 0.000 2.33e-10 5.85e-10
===================================================================================
Ljung-Box (L1) (Q): 60.24 Jarque-Bera (JB): 1334882.31
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.11 Skew: -3.81
Prob(H) (two-sided): 0.00 Kurtosis: 202.35
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 5.73e+20. Standard errors may be unstable.
ARIMA order: (1, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04321, saving model to LSTM8.h5
48/48 - 3s - loss: 1.3590 - val_loss: 0.0432 - lr: 0.0010 - 3s/epoch - 69ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.04321
48/48 - 0s - loss: 1.2809 - val_loss: 0.0457 - lr: 0.0010 - 215ms/epoch - 4ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04321
48/48 - 0s - loss: 1.2036 - val_loss: 0.0497 - lr: 0.0010 - 228ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04321
48/48 - 0s - loss: 1.1161 - val_loss: 0.0545 - lr: 0.0010 - 221ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.04321
48/48 - 0s - loss: 1.0402 - val_loss: 0.0591 - lr: 0.0010 - 223ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9788 - val_loss: 0.0637 - lr: 0.0010 - 226ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9461 - val_loss: 0.0642 - lr: 1.0000e-04 - 228ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9414 - val_loss: 0.0646 - lr: 1.0000e-04 - 215ms/epoch - 4ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9368 - val_loss: 0.0651 - lr: 1.0000e-04 - 227ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9323 - val_loss: 0.0656 - lr: 1.0000e-04 - 215ms/epoch - 4ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9278 - val_loss: 0.0662 - lr: 1.0000e-04 - 218ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9250 - val_loss: 0.0662 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9246 - val_loss: 0.0663 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9242 - val_loss: 0.0663 - lr: 1.0000e-05 - 235ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9237 - val_loss: 0.0664 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9233 - val_loss: 0.0664 - lr: 1.0000e-05 - 234ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9228 - val_loss: 0.0665 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9224 - val_loss: 0.0666 - lr: 1.0000e-05 - 227ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9219 - val_loss: 0.0666 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9215 - val_loss: 0.0667 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9210 - val_loss: 0.0668 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9206 - val_loss: 0.0668 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9201 - val_loss: 0.0669 - lr: 1.0000e-05 - 236ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9197 - val_loss: 0.0670 - lr: 1.0000e-05 - 229ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9192 - val_loss: 0.0670 - lr: 1.0000e-05 - 231ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9188 - val_loss: 0.0671 - lr: 1.0000e-05 - 215ms/epoch - 4ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9183 - val_loss: 0.0672 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9179 - val_loss: 0.0673 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9174 - val_loss: 0.0673 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9170 - val_loss: 0.0674 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9165 - val_loss: 0.0675 - lr: 1.0000e-05 - 227ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9161 - val_loss: 0.0676 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9156 - val_loss: 0.0676 - lr: 1.0000e-05 - 234ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9152 - val_loss: 0.0677 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9147 - val_loss: 0.0678 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9143 - val_loss: 0.0679 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9138 - val_loss: 0.0679 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9134 - val_loss: 0.0680 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9129 - val_loss: 0.0681 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9125 - val_loss: 0.0682 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9120 - val_loss: 0.0683 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9115 - val_loss: 0.0684 - lr: 1.0000e-05 - 241ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9111 - val_loss: 0.0684 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9106 - val_loss: 0.0685 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9102 - val_loss: 0.0686 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9098 - val_loss: 0.0687 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9093 - val_loss: 0.0688 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9089 - val_loss: 0.0689 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9084 - val_loss: 0.0690 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9080 - val_loss: 0.0691 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.04321
48/48 - 0s - loss: 0.9075 - val_loss: 0.0691 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 27.593274833467863
RMSE: 5.252930118844897
MAPE: 4.117405060238624
EMA
EMA([input_arrays], [timeperiod=30])
Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
51
Working on EMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.775, Time=2.28 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=3.86 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16801.081, Time=8.74 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.08 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-14532.068, Time=7.00 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15610.472, Time=11.60 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16103.302, Time=12.94 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17030.021, Time=4.12 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.614, Time=2.92 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.441, Time=5.84 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=17.14 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17003.984, Time=3.03 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-16998.666, Time=3.23 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 87.797 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.720
Date: Sun, 12 Dec 2021 AIC -17087.441
Time: 15:55:29 BIC -16965.479
Sample: 0 HQIC -17040.602
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.316e-10 9.87e-21 -2.35e+10 0.000 -2.32e-10 -2.32e-10
x2 -2.308e-10 9.85e-21 -2.34e+10 0.000 -2.31e-10 -2.31e-10
x3 -2.324e-10 9.88e-21 -2.35e+10 0.000 -2.32e-10 -2.32e-10
x4 1.0000 9.87e-21 1.01e+20 0.000 1.000 1.000
x5 -2.106e-10 9.41e-21 -2.24e+10 0.000 -2.11e-10 -2.11e-10
x6 -7.996e-10 1.74e-20 -4.59e+10 0.000 -8e-10 -8e-10
x7 -2.295e-10 9.82e-21 -2.34e+10 0.000 -2.29e-10 -2.29e-10
x8 -2.244e-10 9.71e-21 -2.31e+10 0.000 -2.24e-10 -2.24e-10
x9 -1.166e-11 1.98e-21 -5.9e+09 0.000 -1.17e-11 -1.17e-11
x10 -4.453e-11 4.22e-21 -1.06e+10 0.000 -4.45e-11 -4.45e-11
x11 -2.219e-10 9.65e-21 -2.3e+10 0.000 -2.22e-10 -2.22e-10
x12 -2.264e-10 9.76e-21 -2.32e+10 0.000 -2.26e-10 -2.26e-10
x13 -2.315e-10 9.87e-21 -2.35e+10 0.000 -2.31e-10 -2.31e-10
x14 -1.766e-09 2.73e-20 -6.48e+10 0.000 -1.77e-09 -1.77e-09
x15 -2.167e-10 9.37e-21 -2.31e+10 0.000 -2.17e-10 -2.17e-10
x16 -5.232e-10 1.49e-20 -3.52e+10 0.000 -5.23e-10 -5.23e-10
x17 -2.147e-10 9.48e-21 -2.27e+10 0.000 -2.15e-10 -2.15e-10
x18 -3.791e-11 3.96e-21 -9.56e+09 0.000 -3.79e-11 -3.79e-11
x19 -2.597e-10 1.05e-20 -2.48e+10 0.000 -2.6e-10 -2.6e-10
x20 -2.417e-10 1e-20 -2.41e+10 0.000 -2.42e-10 -2.42e-10
x21 -4.823e-10 1.4e-20 -3.44e+10 0.000 -4.82e-10 -4.82e-10
ar.L1 -0.4923 1.46e-22 -3.38e+21 0.000 -0.492 -0.492
ar.L2 -0.1923 8.47e-23 -2.27e+21 0.000 -0.192 -0.192
ar.L3 -0.0462 4.02e-23 -1.15e+21 0.000 -0.046 -0.046
ma.L1 -0.7077 3.31e-22 -2.14e+21 0.000 -0.708 -0.708
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 54.09 Jarque-Bera (JB): 4207353.17
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.48
Prob(H) (two-sided): 0.00 Kurtosis: 357.00
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 3.15e+40. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.05220, saving model to LSTM8.h5
16/16 - 4s - loss: 1.3970 - val_loss: 0.0522 - lr: 0.0010 - 4s/epoch - 233ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.3477 - val_loss: 0.0530 - lr: 0.0010 - 86ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.2993 - val_loss: 0.0538 - lr: 0.0010 - 87ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.2538 - val_loss: 0.0549 - lr: 0.0010 - 91ms/epoch - 6ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.2129 - val_loss: 0.0564 - lr: 0.0010 - 94ms/epoch - 6ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1761 - val_loss: 0.0581 - lr: 0.0010 - 92ms/epoch - 6ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1538 - val_loss: 0.0583 - lr: 1.0000e-04 - 89ms/epoch - 6ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1505 - val_loss: 0.0584 - lr: 1.0000e-04 - 94ms/epoch - 6ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1473 - val_loss: 0.0586 - lr: 1.0000e-04 - 83ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1441 - val_loss: 0.0588 - lr: 1.0000e-04 - 89ms/epoch - 6ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1409 - val_loss: 0.0589 - lr: 1.0000e-04 - 82ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1388 - val_loss: 0.0589 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1385 - val_loss: 0.0590 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1382 - val_loss: 0.0590 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1379 - val_loss: 0.0590 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1376 - val_loss: 0.0590 - lr: 1.0000e-05 - 92ms/epoch - 6ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1373 - val_loss: 0.0590 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1369 - val_loss: 0.0590 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1366 - val_loss: 0.0591 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1363 - val_loss: 0.0591 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1360 - val_loss: 0.0591 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1357 - val_loss: 0.0591 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1353 - val_loss: 0.0591 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1350 - val_loss: 0.0591 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1347 - val_loss: 0.0592 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1344 - val_loss: 0.0592 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1340 - val_loss: 0.0592 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1337 - val_loss: 0.0592 - lr: 1.0000e-05 - 98ms/epoch - 6ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1334 - val_loss: 0.0592 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1331 - val_loss: 0.0592 - lr: 1.0000e-05 - 89ms/epoch - 6ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1328 - val_loss: 0.0593 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1324 - val_loss: 0.0593 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1321 - val_loss: 0.0593 - lr: 1.0000e-05 - 84ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1318 - val_loss: 0.0593 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1315 - val_loss: 0.0593 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1311 - val_loss: 0.0593 - lr: 1.0000e-05 - 83ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1308 - val_loss: 0.0593 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1305 - val_loss: 0.0594 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1302 - val_loss: 0.0594 - lr: 1.0000e-05 - 93ms/epoch - 6ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1298 - val_loss: 0.0594 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1295 - val_loss: 0.0594 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1292 - val_loss: 0.0594 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1289 - val_loss: 0.0594 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1285 - val_loss: 0.0595 - lr: 1.0000e-05 - 89ms/epoch - 6ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1282 - val_loss: 0.0595 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1279 - val_loss: 0.0595 - lr: 1.0000e-05 - 90ms/epoch - 6ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1275 - val_loss: 0.0595 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1272 - val_loss: 0.0595 - lr: 1.0000e-05 - 89ms/epoch - 6ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1269 - val_loss: 0.0595 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1266 - val_loss: 0.0596 - lr: 1.0000e-05 - 85ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.05220
16/16 - 0s - loss: 1.1262 - val_loss: 0.0596 - lr: 1.0000e-05 - 91ms/epoch - 6ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 27.593274833467863
RMSE: 5.252930118844897
MAPE: 4.117405060238624
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 36.00024912834101
RMSE: 6.000020760659167
MAPE: 4.70426166066095
WMA
WMA([input_arrays], [timeperiod=30])
Weighted Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
49
Working on WMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-14480.432, Time=9.06 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-15747.905, Time=6.31 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15116.389, Time=6.99 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-13532.115, Time=7.99 sec
ARIMA(0,3,0)(0,0,0)[0] intercept : AIC=-13619.624, Time=5.40 sec
Best model: ARIMA(0,3,0)(0,0,0)[0]
Total fit time: 35.768 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(0, 3, 0) Log Likelihood 7895.952
Date: Sun, 12 Dec 2021 AIC -15747.905
Time: 16:03:44 BIC -15644.706
Sample: 0 HQIC -15708.272
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 3.384e-05 1.9e-05 1.778 0.075 -3.47e-06 7.12e-05
x2 3.379e-05 1.84e-05 1.832 0.067 -2.35e-06 6.99e-05
x3 3.388e-05 4.34e-05 0.781 0.435 -5.12e-05 0.000
x4 1.0000 4.12e-06 2.43e+05 0.000 1.000 1.000
x5 3.227e-05 3.52e-06 9.163 0.000 2.54e-05 3.92e-05
x6 5.559e-05 6.75e-05 0.823 0.410 -7.67e-05 0.000
x7 3.369e-05 2.38e-05 1.415 0.157 -1.3e-05 8.03e-05
x8 0.0023 2.6e-05 86.661 0.000 0.002 0.002
x9 -8.72e-06 7.51e-07 -11.610 0.000 -1.02e-05 -7.25e-06
x10 -0.0023 3.33e-05 -67.770 0.000 -0.002 -0.002
x11 0.0093 2.8e-05 333.459 0.000 0.009 0.009
x12 -0.0118 2.37e-05 -498.171 0.000 -0.012 -0.012
x13 3.382e-05 1.49e-05 2.273 0.023 4.66e-06 6.3e-05
x14 9.271e-05 6.21e-05 1.493 0.135 -2.9e-05 0.000
x15 3.096e-05 1.92e-05 1.614 0.106 -6.63e-06 6.86e-05
x16 5.52e-05 7.17e-05 0.770 0.441 -8.53e-05 0.000
x17 3.38e-05 3.2e-05 1.056 0.291 -2.89e-05 9.65e-05
x18 -6.715e-06 8.34e-05 -0.081 0.936 -0.000 0.000
x19 3.428e-05 2.07e-05 1.654 0.098 -6.34e-06 7.49e-05
x20 -8.089e-06 9.55e-05 -0.085 0.933 -0.000 0.000
x21 4.255e-05 0.000 0.094 0.925 -0.001 0.001
sigma2 2.581e-10 7.87e-11 3.280 0.001 1.04e-10 4.12e-10
===================================================================================
Ljung-Box (L1) (Q): 362.92 Jarque-Bera (JB): 5047564.68
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.01 Skew: -11.23
Prob(H) (two-sided): 0.00 Kurtosis: 390.28
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.75e+20. Standard errors may be unstable.
ARIMA order: (0, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.05515, saving model to LSTM8.h5
17/17 - 4s - loss: 1.3965 - val_loss: 0.0551 - lr: 0.0010 - 4s/epoch - 221ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.3184 - val_loss: 0.0556 - lr: 0.0010 - 90ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.2462 - val_loss: 0.0561 - lr: 0.0010 - 90ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.1850 - val_loss: 0.0570 - lr: 0.0010 - 99ms/epoch - 6ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.1360 - val_loss: 0.0583 - lr: 0.0010 - 90ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0933 - val_loss: 0.0598 - lr: 0.0010 - 94ms/epoch - 6ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0660 - val_loss: 0.0600 - lr: 1.0000e-04 - 96ms/epoch - 6ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0620 - val_loss: 0.0601 - lr: 1.0000e-04 - 93ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0580 - val_loss: 0.0603 - lr: 1.0000e-04 - 94ms/epoch - 6ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0541 - val_loss: 0.0605 - lr: 1.0000e-04 - 95ms/epoch - 6ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0503 - val_loss: 0.0606 - lr: 1.0000e-04 - 94ms/epoch - 6ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0477 - val_loss: 0.0607 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0473 - val_loss: 0.0607 - lr: 1.0000e-05 - 109ms/epoch - 6ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0469 - val_loss: 0.0607 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0465 - val_loss: 0.0607 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0462 - val_loss: 0.0607 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0458 - val_loss: 0.0607 - lr: 1.0000e-05 - 98ms/epoch - 6ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0454 - val_loss: 0.0608 - lr: 1.0000e-05 - 90ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0450 - val_loss: 0.0608 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0446 - val_loss: 0.0608 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0443 - val_loss: 0.0608 - lr: 1.0000e-05 - 88ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0439 - val_loss: 0.0608 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0435 - val_loss: 0.0608 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0431 - val_loss: 0.0609 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0427 - val_loss: 0.0609 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0423 - val_loss: 0.0609 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0420 - val_loss: 0.0609 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0416 - val_loss: 0.0609 - lr: 1.0000e-05 - 86ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0412 - val_loss: 0.0609 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0408 - val_loss: 0.0610 - lr: 1.0000e-05 - 87ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0404 - val_loss: 0.0610 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0400 - val_loss: 0.0610 - lr: 1.0000e-05 - 94ms/epoch - 6ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0396 - val_loss: 0.0610 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0393 - val_loss: 0.0610 - lr: 1.0000e-05 - 89ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0389 - val_loss: 0.0610 - lr: 1.0000e-05 - 92ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0385 - val_loss: 0.0611 - lr: 1.0000e-05 - 91ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0381 - val_loss: 0.0611 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0377 - val_loss: 0.0611 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0373 - val_loss: 0.0611 - lr: 1.0000e-05 - 88ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0369 - val_loss: 0.0611 - lr: 1.0000e-05 - 90ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0366 - val_loss: 0.0612 - lr: 1.0000e-05 - 95ms/epoch - 6ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0362 - val_loss: 0.0612 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0358 - val_loss: 0.0612 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0354 - val_loss: 0.0612 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0350 - val_loss: 0.0612 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0346 - val_loss: 0.0612 - lr: 1.0000e-05 - 101ms/epoch - 6ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0343 - val_loss: 0.0613 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0339 - val_loss: 0.0613 - lr: 1.0000e-05 - 99ms/epoch - 6ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0335 - val_loss: 0.0613 - lr: 1.0000e-05 - 96ms/epoch - 6ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0331 - val_loss: 0.0613 - lr: 1.0000e-05 - 93ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.05515
17/17 - 0s - loss: 1.0327 - val_loss: 0.0613 - lr: 1.0000e-05 - 97ms/epoch - 6ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 27.593274833467863
RMSE: 5.252930118844897
MAPE: 4.117405060238624
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 36.00024912834101
RMSE: 6.000020760659167
MAPE: 4.70426166066095
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 41.43257052147296
RMSE: 6.4368136932393005
MAPE: 5.073793611210466
DEMA
DEMA([input_arrays], [timeperiod=30])
Double Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
89
Working on DEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.774, Time=2.25 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=3.92 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15590.302, Time=6.94 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.27 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15269.503, Time=6.70 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-16414.961, Time=8.36 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16878.396, Time=9.51 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17030.019, Time=4.41 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.613, Time=3.00 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.441, Time=6.15 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=14.61 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17003.985, Time=3.09 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-16998.665, Time=3.57 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 77.793 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.721
Date: Sun, 12 Dec 2021 AIC -17087.441
Time: 16:05:33 BIC -16965.479
Sample: 0 HQIC -17040.603
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.799e-10 1.43e-20 -1.96e+10 0.000 -2.8e-10 -2.8e-10
x2 -2.817e-10 1.43e-20 -1.97e+10 0.000 -2.82e-10 -2.82e-10
x3 -2.805e-10 1.43e-20 -1.96e+10 0.000 -2.8e-10 -2.8e-10
x4 1.0000 1.43e-20 7e+19 0.000 1.000 1.000
x5 -2.597e-10 1.37e-20 -1.89e+10 0.000 -2.6e-10 -2.6e-10
x6 -1.388e-09 3.12e-20 -4.45e+10 0.000 -1.39e-09 -1.39e-09
x7 -2.789e-10 1.42e-20 -1.96e+10 0.000 -2.79e-10 -2.79e-10
x8 -2.76e-10 1.42e-20 -1.95e+10 0.000 -2.76e-10 -2.76e-10
x9 -2.216e-12 3.53e-22 -6.28e+09 0.000 -2.22e-12 -2.22e-12
x10 -1.345e-10 9.82e-21 -1.37e+10 0.000 -1.34e-10 -1.34e-10
x11 -2.898e-10 1.45e-20 -2e+10 0.000 -2.9e-10 -2.9e-10
x12 -2.602e-10 1.38e-20 -1.89e+10 0.000 -2.6e-10 -2.6e-10
x13 -2.807e-10 1.43e-20 -1.96e+10 0.000 -2.81e-10 -2.81e-10
x14 -1.87e-09 3.69e-20 -5.07e+10 0.000 -1.87e-09 -1.87e-09
x15 -2.726e-10 1.43e-20 -1.9e+10 0.000 -2.73e-10 -2.73e-10
x16 -7.915e-11 7.68e-21 -1.03e+10 0.000 -7.92e-11 -7.92e-11
x17 -2.606e-10 1.33e-20 -1.96e+10 0.000 -2.61e-10 -2.61e-10
x18 -6.408e-10 2.16e-20 -2.97e+10 0.000 -6.41e-10 -6.41e-10
x19 -2.881e-10 1.46e-20 -1.98e+10 0.000 -2.88e-10 -2.88e-10
x20 -4.337e-10 1.78e-20 -2.44e+10 0.000 -4.34e-10 -4.34e-10
x21 -4.549e-10 1.79e-20 -2.55e+10 0.000 -4.55e-10 -4.55e-10
ar.L1 -0.4923 1.46e-22 -3.38e+21 0.000 -0.492 -0.492
ar.L2 -0.1923 8.47e-23 -2.27e+21 0.000 -0.192 -0.192
ar.L3 -0.0461 4.02e-23 -1.15e+21 0.000 -0.046 -0.046
ma.L1 -0.7078 3.31e-22 -2.14e+21 0.000 -0.708 -0.708
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 55.07 Jarque-Bera (JB): 4171695.82
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.26
Prob(H) (two-sided): 0.00 Kurtosis: 355.51
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 2.62e+41. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.03617, saving model to LSTM8.h5
10/10 - 3s - loss: 1.4293 - val_loss: 0.0362 - lr: 0.0010 - 3s/epoch - 333ms/step
Epoch 2/500
Epoch 00002: val_loss improved from 0.03617 to 0.03594, saving model to LSTM8.h5
10/10 - 0s - loss: 1.3853 - val_loss: 0.0359 - lr: 0.0010 - 81ms/epoch - 8ms/step
Epoch 3/500
Epoch 00003: val_loss improved from 0.03594 to 0.03580, saving model to LSTM8.h5
10/10 - 0s - loss: 1.3458 - val_loss: 0.0358 - lr: 0.0010 - 73ms/epoch - 7ms/step
Epoch 4/500
Epoch 00004: val_loss improved from 0.03580 to 0.03575, saving model to LSTM8.h5
10/10 - 0s - loss: 1.3096 - val_loss: 0.0357 - lr: 0.0010 - 76ms/epoch - 8ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.2761 - val_loss: 0.0358 - lr: 0.0010 - 61ms/epoch - 6ms/step
Epoch 6/500
Epoch 00006: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.2446 - val_loss: 0.0359 - lr: 0.0010 - 57ms/epoch - 6ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.2145 - val_loss: 0.0361 - lr: 0.0010 - 60ms/epoch - 6ms/step
Epoch 8/500
Epoch 00008: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00008: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1853 - val_loss: 0.0364 - lr: 0.0010 - 58ms/epoch - 6ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1653 - val_loss: 0.0364 - lr: 1.0000e-04 - 63ms/epoch - 6ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1625 - val_loss: 0.0364 - lr: 1.0000e-04 - 64ms/epoch - 6ms/step
Epoch 11/500
Epoch 00011: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1596 - val_loss: 0.0365 - lr: 1.0000e-04 - 70ms/epoch - 7ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1568 - val_loss: 0.0365 - lr: 1.0000e-04 - 66ms/epoch - 7ms/step
Epoch 13/500
Epoch 00013: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00013: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1540 - val_loss: 0.0365 - lr: 1.0000e-04 - 66ms/epoch - 7ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1520 - val_loss: 0.0365 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1518 - val_loss: 0.0365 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step
Epoch 16/500
Epoch 00016: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1515 - val_loss: 0.0365 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1512 - val_loss: 0.0366 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 18/500
Epoch 00018: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00018: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1509 - val_loss: 0.0366 - lr: 1.0000e-05 - 70ms/epoch - 7ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1506 - val_loss: 0.0366 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1503 - val_loss: 0.0366 - lr: 1.0000e-05 - 67ms/epoch - 7ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1500 - val_loss: 0.0366 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1498 - val_loss: 0.0366 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1495 - val_loss: 0.0366 - lr: 1.0000e-05 - 62ms/epoch - 6ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1492 - val_loss: 0.0366 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1489 - val_loss: 0.0366 - lr: 1.0000e-05 - 66ms/epoch - 7ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1486 - val_loss: 0.0366 - lr: 1.0000e-05 - 72ms/epoch - 7ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1483 - val_loss: 0.0366 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1480 - val_loss: 0.0366 - lr: 1.0000e-05 - 64ms/epoch - 6ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1477 - val_loss: 0.0366 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1474 - val_loss: 0.0366 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1471 - val_loss: 0.0366 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1468 - val_loss: 0.0366 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1466 - val_loss: 0.0366 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1463 - val_loss: 0.0366 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1460 - val_loss: 0.0366 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1457 - val_loss: 0.0366 - lr: 1.0000e-05 - 67ms/epoch - 7ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1454 - val_loss: 0.0366 - lr: 1.0000e-05 - 60ms/epoch - 6ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1451 - val_loss: 0.0366 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1448 - val_loss: 0.0366 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1445 - val_loss: 0.0366 - lr: 1.0000e-05 - 69ms/epoch - 7ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1442 - val_loss: 0.0367 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1439 - val_loss: 0.0367 - lr: 1.0000e-05 - 65ms/epoch - 6ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1436 - val_loss: 0.0367 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1433 - val_loss: 0.0367 - lr: 1.0000e-05 - 74ms/epoch - 7ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1430 - val_loss: 0.0367 - lr: 1.0000e-05 - 67ms/epoch - 7ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1427 - val_loss: 0.0367 - lr: 1.0000e-05 - 64ms/epoch - 6ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1424 - val_loss: 0.0367 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1421 - val_loss: 0.0367 - lr: 1.0000e-05 - 68ms/epoch - 7ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1418 - val_loss: 0.0367 - lr: 1.0000e-05 - 63ms/epoch - 6ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1416 - val_loss: 0.0367 - lr: 1.0000e-05 - 58ms/epoch - 6ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1413 - val_loss: 0.0367 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step
Epoch 52/500
Epoch 00052: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1410 - val_loss: 0.0367 - lr: 1.0000e-05 - 62ms/epoch - 6ms/step
Epoch 53/500
Epoch 00053: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1407 - val_loss: 0.0367 - lr: 1.0000e-05 - 59ms/epoch - 6ms/step
Epoch 54/500
Epoch 00054: val_loss did not improve from 0.03575
10/10 - 0s - loss: 1.1404 - val_loss: 0.0367 - lr: 1.0000e-05 - 61ms/epoch - 6ms/step
Epoch 00054: early stopping
SMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 27.593274833467863
RMSE: 5.252930118844897
MAPE: 4.117405060238624
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 36.00024912834101
RMSE: 6.000020760659167
MAPE: 4.70426166066095
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 41.43257052147296
RMSE: 6.4368136932393005
MAPE: 5.073793611210466
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.15% Accuracy
MSE: 181.24518187853474
RMSE: 13.462733076108087
MAPE: 12.35792879577932
KAMA
KAMA([input_arrays], [timeperiod=30])
Kaufman Adaptive Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
18
Working on KAMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.902, Time=2.11 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.593, Time=3.85 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16796.316, Time=7.98 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.593, Time=5.53 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-17004.193, Time=2.45 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15176.063, Time=9.99 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-16873.638, Time=9.63 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17008.756, Time=2.46 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.764, Time=3.05 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-15723.849, Time=14.32 sec
ARIMA(2,3,0)(0,0,0)[0] intercept : AIC=-17006.756, Time=2.85 sec
Best model: ARIMA(2,3,0)(0,0,0)[0]
Total fit time: 64.230 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(2, 3, 0) Log Likelihood 8528.378
Date: Sun, 12 Dec 2021 AIC -17008.756
Time: 16:14:14 BIC -16896.176
Sample: 0 HQIC -16965.520
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.24e-15 7.41e-26 -3.02e+10 0.000 -2.24e-15 -2.24e-15
x2 8.461e-16 6.6e-26 1.28e+10 0.000 8.46e-16 8.46e-16
x3 4.901e-16 6.89e-26 7.11e+09 0.000 4.9e-16 4.9e-16
x4 1.0000 6.96e-26 1.44e+25 0.000 1.000 1.000
x5 5.931e-15 6.61e-26 8.97e+10 0.000 5.93e-15 5.93e-15
x6 -1.05e-15 1.5e-25 -7e+09 0.000 -1.05e-15 -1.05e-15
x7 1.439e-15 6.87e-26 2.1e+10 0.000 1.44e-15 1.44e-15
x8 -1.25e-15 6.8e-26 -1.84e+10 0.000 -1.25e-15 -1.25e-15
x9 -9.356e-17 8.97e-27 -1.04e+10 0.000 -9.36e-17 -9.36e-17
x10 -1.145e-16 2.88e-26 -3.98e+09 0.000 -1.15e-16 -1.15e-16
x11 -2.036e-16 6.8e-26 -3e+09 0.000 -2.04e-16 -2.04e-16
x12 5.951e-16 6.76e-26 8.81e+09 0.000 5.95e-16 5.95e-16
x13 -6.117e-17 6.94e-26 -8.81e+08 0.000 -6.12e-17 -6.12e-17
x14 1.167e-15 1.99e-25 5.85e+09 0.000 1.17e-15 1.17e-15
x15 -4.274e-14 6.99e-26 -6.11e+11 0.000 -4.27e-14 -4.27e-14
x16 2.262e-14 8.56e-26 2.64e+11 0.000 2.26e-14 2.26e-14
x17 3.384e-14 6.46e-26 5.24e+11 0.000 3.38e-14 3.38e-14
x18 9.894e-16 5.8e-26 1.71e+10 0.000 9.89e-16 9.89e-16
x19 4.115e-14 7.75e-26 5.31e+11 0.000 4.12e-14 4.12e-14
x20 -2.176e-15 9.49e-26 -2.29e+10 0.000 -2.18e-15 -2.18e-15
x21 -7.755e-17 4.63e-26 -1.67e+09 0.000 -7.75e-17 -7.75e-17
ar.L1 -0.9988 9.76e-22 -1.02e+21 0.000 -0.999 -0.999
ar.L2 -0.4972 4.07e-23 -1.22e+22 0.000 -0.497 -0.497
sigma2 1e-10 6.99e-11 1.432 0.152 -3.69e-11 2.37e-10
===================================================================================
Ljung-Box (L1) (Q): 31.54 Jarque-Bera (JB): 2432532.03
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: -0.15
Prob(H) (two-sided): 0.00 Kurtosis: 272.30
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 7.19e+48. Standard errors may be unstable.
ARIMA order: (2, 3, 0)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04342, saving model to LSTM8.h5
45/45 - 4s - loss: 1.3084 - val_loss: 0.0434 - lr: 0.0010 - 4s/epoch - 78ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.04342
45/45 - 0s - loss: 1.0770 - val_loss: 0.0462 - lr: 0.0010 - 212ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.9119 - val_loss: 0.0491 - lr: 0.0010 - 204ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.8332 - val_loss: 0.0518 - lr: 0.0010 - 227ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7877 - val_loss: 0.0544 - lr: 0.0010 - 218ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7566 - val_loss: 0.0570 - lr: 0.0010 - 207ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7413 - val_loss: 0.0573 - lr: 1.0000e-04 - 211ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7391 - val_loss: 0.0575 - lr: 1.0000e-04 - 211ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7368 - val_loss: 0.0578 - lr: 1.0000e-04 - 215ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7346 - val_loss: 0.0581 - lr: 1.0000e-04 - 216ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7323 - val_loss: 0.0584 - lr: 1.0000e-04 - 219ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7309 - val_loss: 0.0584 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7307 - val_loss: 0.0585 - lr: 1.0000e-05 - 214ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7304 - val_loss: 0.0585 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7302 - val_loss: 0.0586 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7299 - val_loss: 0.0586 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7297 - val_loss: 0.0586 - lr: 1.0000e-05 - 215ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7294 - val_loss: 0.0587 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7292 - val_loss: 0.0587 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7289 - val_loss: 0.0587 - lr: 1.0000e-05 - 211ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7287 - val_loss: 0.0588 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7284 - val_loss: 0.0588 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7281 - val_loss: 0.0589 - lr: 1.0000e-05 - 232ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7279 - val_loss: 0.0589 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7276 - val_loss: 0.0590 - lr: 1.0000e-05 - 208ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7273 - val_loss: 0.0590 - lr: 1.0000e-05 - 201ms/epoch - 4ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7271 - val_loss: 0.0590 - lr: 1.0000e-05 - 231ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7268 - val_loss: 0.0591 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7265 - val_loss: 0.0591 - lr: 1.0000e-05 - 210ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7262 - val_loss: 0.0592 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7260 - val_loss: 0.0592 - lr: 1.0000e-05 - 208ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7257 - val_loss: 0.0593 - lr: 1.0000e-05 - 214ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7254 - val_loss: 0.0593 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7251 - val_loss: 0.0594 - lr: 1.0000e-05 - 215ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7248 - val_loss: 0.0594 - lr: 1.0000e-05 - 215ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7246 - val_loss: 0.0595 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7243 - val_loss: 0.0595 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7240 - val_loss: 0.0596 - lr: 1.0000e-05 - 212ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7237 - val_loss: 0.0596 - lr: 1.0000e-05 - 205ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7234 - val_loss: 0.0597 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7231 - val_loss: 0.0598 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7228 - val_loss: 0.0598 - lr: 1.0000e-05 - 215ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7226 - val_loss: 0.0599 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7223 - val_loss: 0.0599 - lr: 1.0000e-05 - 215ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7220 - val_loss: 0.0600 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7217 - val_loss: 0.0600 - lr: 1.0000e-05 - 227ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7214 - val_loss: 0.0601 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7211 - val_loss: 0.0602 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7208 - val_loss: 0.0602 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7205 - val_loss: 0.0603 - lr: 1.0000e-05 - 237ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.04342
45/45 - 0s - loss: 0.7203 - val_loss: 0.0603 - lr: 1.0000e-05 - 227ms/epoch - 5ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 27.593274833467863
RMSE: 5.252930118844897
MAPE: 4.117405060238624
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 36.00024912834101
RMSE: 6.000020760659167
MAPE: 4.70426166066095
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 41.43257052147296
RMSE: 6.4368136932393005
MAPE: 5.073793611210466
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.15% Accuracy
MSE: 181.24518187853474
RMSE: 13.462733076108087
MAPE: 12.35792879577932
KAMA
Prediction vs Close: 56.72% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 27.433326731723586
RMSE: 5.237683336335214
MAPE: 4.154161853007022
MIDPOINT
MIDPOINT([input_arrays], [timeperiod=14])
MidPoint over period (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 14
Outputs:
real
14
Working on MIDPOINT predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-17005.753, Time=2.34 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14574.592, Time=3.97 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16288.639, Time=11.00 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-14572.592, Time=5.50 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15275.254, Time=7.01 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15486.751, Time=12.68 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=48.000, Time=0.54 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-17008.491, Time=2.31 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-17004.554, Time=3.03 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-17087.445, Time=6.14 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=-15686.421, Time=10.07 sec
ARIMA(2,3,2)(0,0,0)[0] : AIC=-17030.168, Time=13.88 sec
ARIMA(3,3,1)(0,0,0)[0] intercept : AIC=-15138.715, Time=14.55 sec
Best model: ARIMA(3,3,1)(0,0,0)[0]
Total fit time: 93.028 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(3, 3, 1) Log Likelihood 8569.722
Date: Sun, 12 Dec 2021 AIC -17087.445
Time: 16:16:53 BIC -16965.483
Sample: 0 HQIC -17040.607
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 -2.14e-10 1.09e-20 -1.96e+10 0.000 -2.14e-10 -2.14e-10
x2 -2.126e-10 1.13e-20 -1.88e+10 0.000 -2.13e-10 -2.13e-10
x3 -2.175e-10 1.06e-20 -2.04e+10 0.000 -2.17e-10 -2.17e-10
x4 1.0000 1.1e-20 9.11e+19 0.000 1.000 1.000
x5 -1.941e-10 1.05e-20 -1.86e+10 0.000 -1.94e-10 -1.94e-10
x6 -4.131e-09 7.64e-20 -5.4e+10 0.000 -4.13e-09 -4.13e-09
x7 -1.965e-10 1.05e-20 -1.86e+10 0.000 -1.96e-10 -1.96e-10
x8 -1.961e-10 1.07e-20 -1.84e+10 0.000 -1.96e-10 -1.96e-10
x9 -1.005e-10 9.12e-22 -1.1e+11 0.000 -1e-10 -1e-10
x10 -1.739e-10 3.37e-21 -5.16e+10 0.000 -1.74e-10 -1.74e-10
x11 -1.941e-10 1.07e-20 -1.82e+10 0.000 -1.94e-10 -1.94e-10
x12 -2.005e-10 1.06e-20 -1.89e+10 0.000 -2e-10 -2e-10
x13 -2.056e-10 1.07e-20 -1.91e+10 0.000 -2.06e-10 -2.06e-10
x14 -1.687e-09 3.15e-20 -5.36e+10 0.000 -1.69e-09 -1.69e-09
x15 -2.365e-10 1.17e-20 -2.01e+10 0.000 -2.36e-10 -2.36e-10
x16 -1.523e-10 9.42e-21 -1.62e+10 0.000 -1.52e-10 -1.52e-10
x17 -1.491e-10 9.33e-21 -1.6e+10 0.000 -1.49e-10 -1.49e-10
x18 -6.404e-10 1.93e-20 -3.32e+10 0.000 -6.4e-10 -6.4e-10
x19 -2.596e-10 1.23e-20 -2.11e+10 0.000 -2.6e-10 -2.6e-10
x20 -6.246e-10 1.91e-20 -3.28e+10 0.000 -6.25e-10 -6.25e-10
x21 -1.953e-09 2.16e-20 -9.04e+10 0.000 -1.95e-09 -1.95e-09
ar.L1 -0.4914 1.46e-22 -3.38e+21 0.000 -0.491 -0.491
ar.L2 -0.1934 8.48e-23 -2.28e+21 0.000 -0.193 -0.193
ar.L3 -0.0491 4.2e-23 -1.17e+21 0.000 -0.049 -0.049
ma.L1 -0.7092 3.33e-22 -2.13e+21 0.000 -0.709 -0.709
sigma2 8.99e-11 6.95e-11 1.293 0.196 -4.64e-11 2.26e-10
===================================================================================
Ljung-Box (L1) (Q): 32.51 Jarque-Bera (JB): 49038.38
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.35 Skew: 1.06
Prob(H) (two-sided): 0.00 Kurtosis: 41.18
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.71e+40. Standard errors may be unstable.
ARIMA order: (3, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04338, saving model to LSTM8.h5
58/58 - 4s - loss: 1.2850 - val_loss: 0.0434 - lr: 0.0010 - 4s/epoch - 67ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.04338
58/58 - 0s - loss: 1.1296 - val_loss: 0.0475 - lr: 0.0010 - 279ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.9824 - val_loss: 0.0530 - lr: 0.0010 - 297ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.8742 - val_loss: 0.0622 - lr: 0.0010 - 297ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.8035 - val_loss: 0.0747 - lr: 0.0010 - 278ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7564 - val_loss: 0.0883 - lr: 0.0010 - 265ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7343 - val_loss: 0.0896 - lr: 1.0000e-04 - 295ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7312 - val_loss: 0.0909 - lr: 1.0000e-04 - 278ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7280 - val_loss: 0.0923 - lr: 1.0000e-04 - 274ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7249 - val_loss: 0.0938 - lr: 1.0000e-04 - 292ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7218 - val_loss: 0.0952 - lr: 1.0000e-04 - 275ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7199 - val_loss: 0.0954 - lr: 1.0000e-05 - 276ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7196 - val_loss: 0.0956 - lr: 1.0000e-05 - 263ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7193 - val_loss: 0.0957 - lr: 1.0000e-05 - 283ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7190 - val_loss: 0.0959 - lr: 1.0000e-05 - 268ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7186 - val_loss: 0.0960 - lr: 1.0000e-05 - 282ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7183 - val_loss: 0.0962 - lr: 1.0000e-05 - 280ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7180 - val_loss: 0.0964 - lr: 1.0000e-05 - 285ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7176 - val_loss: 0.0966 - lr: 1.0000e-05 - 263ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7173 - val_loss: 0.0967 - lr: 1.0000e-05 - 267ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7169 - val_loss: 0.0969 - lr: 1.0000e-05 - 280ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7166 - val_loss: 0.0971 - lr: 1.0000e-05 - 269ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7162 - val_loss: 0.0973 - lr: 1.0000e-05 - 268ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7159 - val_loss: 0.0975 - lr: 1.0000e-05 - 274ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7155 - val_loss: 0.0977 - lr: 1.0000e-05 - 277ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7152 - val_loss: 0.0979 - lr: 1.0000e-05 - 272ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7148 - val_loss: 0.0981 - lr: 1.0000e-05 - 279ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7144 - val_loss: 0.0983 - lr: 1.0000e-05 - 280ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7141 - val_loss: 0.0985 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7137 - val_loss: 0.0987 - lr: 1.0000e-05 - 278ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7133 - val_loss: 0.0989 - lr: 1.0000e-05 - 276ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7130 - val_loss: 0.0991 - lr: 1.0000e-05 - 280ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7126 - val_loss: 0.0993 - lr: 1.0000e-05 - 271ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7122 - val_loss: 0.0996 - lr: 1.0000e-05 - 263ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7118 - val_loss: 0.0998 - lr: 1.0000e-05 - 273ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7115 - val_loss: 0.1000 - lr: 1.0000e-05 - 278ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7111 - val_loss: 0.1002 - lr: 1.0000e-05 - 290ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7107 - val_loss: 0.1004 - lr: 1.0000e-05 - 268ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7103 - val_loss: 0.1006 - lr: 1.0000e-05 - 295ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7100 - val_loss: 0.1009 - lr: 1.0000e-05 - 280ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7096 - val_loss: 0.1011 - lr: 1.0000e-05 - 272ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7092 - val_loss: 0.1013 - lr: 1.0000e-05 - 269ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7088 - val_loss: 0.1015 - lr: 1.0000e-05 - 286ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7085 - val_loss: 0.1018 - lr: 1.0000e-05 - 256ms/epoch - 4ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7081 - val_loss: 0.1020 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7077 - val_loss: 0.1022 - lr: 1.0000e-05 - 267ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7073 - val_loss: 0.1024 - lr: 1.0000e-05 - 273ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7070 - val_loss: 0.1027 - lr: 1.0000e-05 - 256ms/epoch - 4ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7066 - val_loss: 0.1029 - lr: 1.0000e-05 - 258ms/epoch - 4ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7062 - val_loss: 0.1031 - lr: 1.0000e-05 - 285ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.04338
58/58 - 0s - loss: 0.7058 - val_loss: 0.1034 - lr: 1.0000e-05 - 270ms/epoch - 5ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 27.593274833467863
RMSE: 5.252930118844897
MAPE: 4.117405060238624
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 36.00024912834101
RMSE: 6.000020760659167
MAPE: 4.70426166066095
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 41.43257052147296
RMSE: 6.4368136932393005
MAPE: 5.073793611210466
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.15% Accuracy
MSE: 181.24518187853474
RMSE: 13.462733076108087
MAPE: 12.35792879577932
KAMA
Prediction vs Close: 56.72% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 27.433326731723586
RMSE: 5.237683336335214
MAPE: 4.154161853007022
MIDPOINT
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 44.78% Accuracy
MSE: 17.391761137690516
RMSE: 4.170343047962663
MAPE: 3.393694867236689
T3
T3([input_arrays], [timeperiod=5], [vfactor=0.7])
Triple Exponential Moving Average (T3) (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 5
vfactor: 0.7
Outputs:
real
19
Working on T3 predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-16569.270, Time=2.34 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-14511.291, Time=2.50 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-15408.738, Time=7.88 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-15165.005, Time=8.17 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-15595.465, Time=7.31 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-15837.470, Time=9.86 sec
ARIMA(0,3,2)(0,0,0)[0] : AIC=-15491.538, Time=9.06 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-16378.438, Time=2.58 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-16318.604, Time=3.56 sec
ARIMA(1,3,1)(0,0,0)[0] intercept : AIC=-16567.270, Time=2.35 sec
Best model: ARIMA(1,3,1)(0,0,0)[0]
Total fit time: 55.608 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(1, 3, 1) Log Likelihood 8308.635
Date: Sun, 12 Dec 2021 AIC -16569.270
Time: 16:25:22 BIC -16456.690
Sample: 0 HQIC -16526.035
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 1.355e-13 3.43e-05 3.95e-09 1.000 -6.72e-05 6.72e-05
x2 5.009e-14 2.67e-05 1.88e-09 1.000 -5.23e-05 5.23e-05
x3 -9.101e-15 2.19e-05 -4.16e-10 1.000 -4.29e-05 4.29e-05
x4 1.0000 2.97e-05 3.37e+04 0.000 1.000 1.000
x5 3.626e-12 3.2e-05 1.13e-07 1.000 -6.28e-05 6.28e-05
x6 6.879e-17 0.000 5.13e-13 1.000 -0.000 0.000
x7 1.588e-13 4.04e-05 3.93e-09 1.000 -7.92e-05 7.92e-05
x8 -0.0002 9.77e-06 -20.395 0.000 -0.000 -0.000
x9 3.877e-14 0.001 6.24e-11 1.000 -0.001 0.001
x10 -7.41e-05 0.001 -0.129 0.897 -0.001 0.001
x11 0.0003 4.91e-05 6.926 0.000 0.000 0.000
x12 -0.0004 7.27e-05 -5.556 0.000 -0.001 -0.000
x13 -2.679e-14 3.39e-05 -7.9e-10 1.000 -6.65e-05 6.65e-05
x14 2.97e-13 0.000 2.31e-09 1.000 -0.000 0.000
x15 1.602e-12 7.47e-05 2.14e-08 1.000 -0.000 0.000
x16 -8.756e-13 4.29e-05 -2.04e-08 1.000 -8.41e-05 8.41e-05
x17 1.793e-12 6.56e-05 2.74e-08 1.000 -0.000 0.000
x18 -1.019e-13 0.000 -5.54e-10 1.000 -0.000 0.000
x19 -1.077e-12 8.29e-05 -1.3e-08 1.000 -0.000 0.000
x20 1.771e-13 8.45e-05 2.1e-09 1.000 -0.000 0.000
x21 9.233e-16 0.000 1.94e-12 1.000 -0.001 0.001
ar.L1 -0.2857 0.000 -2747.572 0.000 -0.286 -0.285
ma.L1 -0.9142 7.12e-06 -1.28e+05 0.000 -0.914 -0.914
sigma2 1e-10 7e-11 1.429 0.153 -3.71e-11 2.37e-10
===================================================================================
Ljung-Box (L1) (Q): 84.32 Jarque-Bera (JB): 4804295.53
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.87
Prob(H) (two-sided): 0.00 Kurtosis: 381.28
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 2.06e+20. Standard errors may be unstable.
ARIMA order: (1, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.05022, saving model to LSTM8.h5
43/43 - 4s - loss: 1.4172 - val_loss: 0.0502 - lr: 0.0010 - 4s/epoch - 87ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.3610 - val_loss: 0.0527 - lr: 0.0010 - 220ms/epoch - 5ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.2831 - val_loss: 0.0558 - lr: 0.0010 - 208ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.2176 - val_loss: 0.0592 - lr: 0.0010 - 228ms/epoch - 5ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.1635 - val_loss: 0.0629 - lr: 0.0010 - 205ms/epoch - 5ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.1164 - val_loss: 0.0668 - lr: 0.0010 - 211ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0893 - val_loss: 0.0672 - lr: 1.0000e-04 - 198ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0853 - val_loss: 0.0676 - lr: 1.0000e-04 - 212ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0813 - val_loss: 0.0680 - lr: 1.0000e-04 - 211ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0773 - val_loss: 0.0685 - lr: 1.0000e-04 - 197ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0734 - val_loss: 0.0689 - lr: 1.0000e-04 - 206ms/epoch - 5ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0709 - val_loss: 0.0690 - lr: 1.0000e-05 - 195ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0705 - val_loss: 0.0690 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0702 - val_loss: 0.0690 - lr: 1.0000e-05 - 212ms/epoch - 5ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0698 - val_loss: 0.0691 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0694 - val_loss: 0.0691 - lr: 1.0000e-05 - 198ms/epoch - 5ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0690 - val_loss: 0.0692 - lr: 1.0000e-05 - 205ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0686 - val_loss: 0.0692 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0682 - val_loss: 0.0693 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0678 - val_loss: 0.0693 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0674 - val_loss: 0.0694 - lr: 1.0000e-05 - 208ms/epoch - 5ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0670 - val_loss: 0.0694 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0666 - val_loss: 0.0695 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0662 - val_loss: 0.0695 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0658 - val_loss: 0.0696 - lr: 1.0000e-05 - 214ms/epoch - 5ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0654 - val_loss: 0.0696 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0650 - val_loss: 0.0697 - lr: 1.0000e-05 - 215ms/epoch - 5ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0647 - val_loss: 0.0697 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0643 - val_loss: 0.0698 - lr: 1.0000e-05 - 205ms/epoch - 5ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0639 - val_loss: 0.0698 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0635 - val_loss: 0.0699 - lr: 1.0000e-05 - 236ms/epoch - 5ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0631 - val_loss: 0.0699 - lr: 1.0000e-05 - 215ms/epoch - 5ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0627 - val_loss: 0.0700 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0623 - val_loss: 0.0700 - lr: 1.0000e-05 - 195ms/epoch - 5ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0619 - val_loss: 0.0701 - lr: 1.0000e-05 - 200ms/epoch - 5ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0615 - val_loss: 0.0701 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0611 - val_loss: 0.0702 - lr: 1.0000e-05 - 205ms/epoch - 5ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0607 - val_loss: 0.0702 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0603 - val_loss: 0.0703 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0599 - val_loss: 0.0704 - lr: 1.0000e-05 - 199ms/epoch - 5ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0595 - val_loss: 0.0704 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0591 - val_loss: 0.0705 - lr: 1.0000e-05 - 208ms/epoch - 5ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0587 - val_loss: 0.0705 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0583 - val_loss: 0.0706 - lr: 1.0000e-05 - 206ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0579 - val_loss: 0.0706 - lr: 1.0000e-05 - 197ms/epoch - 5ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0576 - val_loss: 0.0707 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0572 - val_loss: 0.0707 - lr: 1.0000e-05 - 209ms/epoch - 5ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0568 - val_loss: 0.0708 - lr: 1.0000e-05 - 207ms/epoch - 5ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0564 - val_loss: 0.0708 - lr: 1.0000e-05 - 213ms/epoch - 5ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0560 - val_loss: 0.0709 - lr: 1.0000e-05 - 197ms/epoch - 5ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.05022
43/43 - 0s - loss: 1.0556 - val_loss: 0.0710 - lr: 1.0000e-05 - 211ms/epoch - 5ms/step
Epoch 00051: early stopping
SMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 47.01% Accuracy
MSE: 27.593274833467863
RMSE: 5.252930118844897
MAPE: 4.117405060238624
EMA
Prediction vs Close: 54.85% Accuracy
Prediction vs Prediction: 49.63% Accuracy
MSE: 36.00024912834101
RMSE: 6.000020760659167
MAPE: 4.70426166066095
WMA
Prediction vs Close: 54.48% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 41.43257052147296
RMSE: 6.4368136932393005
MAPE: 5.073793611210466
DEMA
Prediction vs Close: 52.24% Accuracy
Prediction vs Prediction: 45.15% Accuracy
MSE: 181.24518187853474
RMSE: 13.462733076108087
MAPE: 12.35792879577932
KAMA
Prediction vs Close: 56.72% Accuracy
Prediction vs Prediction: 46.27% Accuracy
MSE: 27.433326731723586
RMSE: 5.237683336335214
MAPE: 4.154161853007022
MIDPOINT
Prediction vs Close: 51.87% Accuracy
Prediction vs Prediction: 44.78% Accuracy
MSE: 17.391761137690516
RMSE: 4.170343047962663
MAPE: 3.393694867236689
T3
Prediction vs Close: 53.73% Accuracy
Prediction vs Prediction: 47.76% Accuracy
MSE: 65.97403618729489
RMSE: 8.122440285240321
MAPE: 6.528054740759965
TEMA
TEMA([input_arrays], [timeperiod=30])
Triple Exponential Moving Average (Overlap Studies)
Inputs:
price: (any ndarray)
Parameters:
timeperiod: 30
Outputs:
real
9
Working on TEMA predictions
parameters used : 808 269
Performing stepwise search to minimize aic
ARIMA(1,3,1)(0,0,0)[0] : AIC=-16493.570, Time=2.72 sec
ARIMA(0,3,0)(0,0,0)[0] : AIC=-15527.581, Time=7.34 sec
ARIMA(1,3,0)(0,0,0)[0] : AIC=-16154.477, Time=7.07 sec
ARIMA(0,3,1)(0,0,0)[0] : AIC=-15134.948, Time=6.40 sec
ARIMA(2,3,1)(0,0,0)[0] : AIC=-16538.454, Time=8.03 sec
ARIMA(2,3,0)(0,0,0)[0] : AIC=-16271.346, Time=2.46 sec
ARIMA(3,3,1)(0,0,0)[0] : AIC=-16350.992, Time=12.60 sec
/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:1906: RuntimeWarning: divide by zero encountered in reciprocal return np.roots(self.polynomial_reduced_ma)**-1
ARIMA(2,3,2)(0,0,0)[0] : AIC=-16200.149, Time=3.43 sec
ARIMA(1,3,2)(0,0,0)[0] : AIC=-16461.809, Time=16.42 sec
ARIMA(3,3,0)(0,0,0)[0] : AIC=-16384.147, Time=3.34 sec
ARIMA(3,3,2)(0,0,0)[0] : AIC=inf, Time=8.64 sec
ARIMA(2,3,1)(0,0,0)[0] intercept : AIC=-15110.164, Time=5.24 sec
Best model: ARIMA(2,3,1)(0,0,0)[0]
Total fit time: 83.700 seconds
SARIMAX Results
==============================================================================
Dep. Variable: y No. Observations: 808
Model: SARIMAX(2, 3, 1) Log Likelihood 8294.227
Date: Sun, 12 Dec 2021 AIC -16538.454
Time: 16:29:48 BIC -16421.183
Sample: 0 HQIC -16493.417
- 808
Covariance Type: opg
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
x1 3.591e-07 0.001 0.000 1.000 -0.002 0.002
x2 3.6e-07 0.002 0.000 1.000 -0.003 0.003
x3 3.611e-07 0.001 0.000 1.000 -0.002 0.002
x4 1.0000 0.000 2628.605 0.000 0.999 1.001
x5 3.432e-07 0.000 0.001 0.999 -0.001 0.001
x6 1.714e-07 4.05e-05 0.004 0.997 -7.91e-05 7.95e-05
x7 3.541e-07 0.001 0.000 1.000 -0.003 0.003
x8 -0.0002 0.000 -1.006 0.315 -0.001 0.000
x9 -7.559e-08 0.000 -0.000 1.000 -0.001 0.001
x10 0.0001 0.000 0.492 0.623 -0.000 0.001
x11 -0.0006 0.000 -2.697 0.007 -0.001 -0.000
x12 0.0005 0.000 1.741 0.082 -5.97e-05 0.001
x13 3.6e-07 0.000 0.002 0.999 -0.000 0.000
x14 1.003e-06 0.001 0.001 0.999 -0.002 0.002
x15 3.506e-07 7.16e-05 0.005 0.996 -0.000 0.000
x16 5.157e-07 0.000 0.005 0.996 -0.000 0.000
x17 3.516e-07 6.59e-05 0.005 0.996 -0.000 0.000
x18 1.166e-07 0.000 0.001 1.000 -0.000 0.000
x19 3.922e-07 7.5e-05 0.005 0.996 -0.000 0.000
x20 -3.64e-07 0.000 -0.002 0.999 -0.000 0.000
x21 4.458e-07 0.000 0.004 0.997 -0.000 0.000
ar.L1 -0.4019 4.12e-05 -9758.484 0.000 -0.402 -0.402
ar.L2 -0.1006 1.58e-05 -6360.873 0.000 -0.101 -0.101
ma.L1 -0.7963 8.45e-06 -9.43e+04 0.000 -0.796 -0.796
sigma2 9.048e-11 7.2e-11 1.257 0.209 -5.06e-11 2.32e-10
===================================================================================
Ljung-Box (L1) (Q): 64.02 Jarque-Bera (JB): 4424775.47
Prob(Q): 0.00 Prob(JB): 0.00
Heteroskedasticity (H): 0.00 Skew: 5.53
Prob(H) (two-sided): 0.00 Kurtosis: 366.04
===================================================================================
Warnings:
[1] Covariance matrix calculated using the outer product of gradients (complex-step).
[2] Covariance matrix is singular or near-singular, with condition number 1.02e+20. Standard errors may be unstable.
ARIMA order: (2, 3, 1)
Epoch 1/500
Epoch 00001: val_loss improved from inf to 0.04701, saving model to LSTM8.h5
90/90 - 4s - loss: 1.2615 - val_loss: 0.0470 - lr: 0.0010 - 4s/epoch - 41ms/step
Epoch 2/500
Epoch 00002: val_loss did not improve from 0.04701
90/90 - 0s - loss: 1.0630 - val_loss: 0.0557 - lr: 0.0010 - 390ms/epoch - 4ms/step
Epoch 3/500
Epoch 00003: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.9138 - val_loss: 0.0651 - lr: 0.0010 - 409ms/epoch - 5ms/step
Epoch 4/500
Epoch 00004: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.8175 - val_loss: 0.0747 - lr: 0.0010 - 395ms/epoch - 4ms/step
Epoch 5/500
Epoch 00005: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.7588 - val_loss: 0.0838 - lr: 0.0010 - 389ms/epoch - 4ms/step
Epoch 6/500
Epoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.
Epoch 00006: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.7204 - val_loss: 0.0927 - lr: 0.0010 - 411ms/epoch - 5ms/step
Epoch 7/500
Epoch 00007: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.7024 - val_loss: 0.0935 - lr: 1.0000e-04 - 430ms/epoch - 5ms/step
Epoch 8/500
Epoch 00008: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6997 - val_loss: 0.0945 - lr: 1.0000e-04 - 427ms/epoch - 5ms/step
Epoch 9/500
Epoch 00009: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6970 - val_loss: 0.0955 - lr: 1.0000e-04 - 414ms/epoch - 5ms/step
Epoch 10/500
Epoch 00010: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6943 - val_loss: 0.0965 - lr: 1.0000e-04 - 419ms/epoch - 5ms/step
Epoch 11/500
Epoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.
Epoch 00011: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6916 - val_loss: 0.0975 - lr: 1.0000e-04 - 400ms/epoch - 4ms/step
Epoch 12/500
Epoch 00012: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6899 - val_loss: 0.0977 - lr: 1.0000e-05 - 411ms/epoch - 5ms/step
Epoch 13/500
Epoch 00013: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6896 - val_loss: 0.0978 - lr: 1.0000e-05 - 399ms/epoch - 4ms/step
Epoch 14/500
Epoch 00014: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6893 - val_loss: 0.0979 - lr: 1.0000e-05 - 394ms/epoch - 4ms/step
Epoch 15/500
Epoch 00015: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6890 - val_loss: 0.0980 - lr: 1.0000e-05 - 397ms/epoch - 4ms/step
Epoch 16/500
Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.
Epoch 00016: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6887 - val_loss: 0.0981 - lr: 1.0000e-05 - 397ms/epoch - 4ms/step
Epoch 17/500
Epoch 00017: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6884 - val_loss: 0.0983 - lr: 1.0000e-05 - 409ms/epoch - 5ms/step
Epoch 18/500
Epoch 00018: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6881 - val_loss: 0.0984 - lr: 1.0000e-05 - 398ms/epoch - 4ms/step
Epoch 19/500
Epoch 00019: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6878 - val_loss: 0.0986 - lr: 1.0000e-05 - 388ms/epoch - 4ms/step
Epoch 20/500
Epoch 00020: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6875 - val_loss: 0.0987 - lr: 1.0000e-05 - 381ms/epoch - 4ms/step
Epoch 21/500
Epoch 00021: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6871 - val_loss: 0.0989 - lr: 1.0000e-05 - 393ms/epoch - 4ms/step
Epoch 22/500
Epoch 00022: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6868 - val_loss: 0.0990 - lr: 1.0000e-05 - 418ms/epoch - 5ms/step
Epoch 23/500
Epoch 00023: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6865 - val_loss: 0.0992 - lr: 1.0000e-05 - 429ms/epoch - 5ms/step
Epoch 24/500
Epoch 00024: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6861 - val_loss: 0.0993 - lr: 1.0000e-05 - 425ms/epoch - 5ms/step
Epoch 25/500
Epoch 00025: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6858 - val_loss: 0.0995 - lr: 1.0000e-05 - 398ms/epoch - 4ms/step
Epoch 26/500
Epoch 00026: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6855 - val_loss: 0.0997 - lr: 1.0000e-05 - 409ms/epoch - 5ms/step
Epoch 27/500
Epoch 00027: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6851 - val_loss: 0.0998 - lr: 1.0000e-05 - 388ms/epoch - 4ms/step
Epoch 28/500
Epoch 00028: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6848 - val_loss: 0.1000 - lr: 1.0000e-05 - 388ms/epoch - 4ms/step
Epoch 29/500
Epoch 00029: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6844 - val_loss: 0.1002 - lr: 1.0000e-05 - 396ms/epoch - 4ms/step
Epoch 30/500
Epoch 00030: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6841 - val_loss: 0.1004 - lr: 1.0000e-05 - 395ms/epoch - 4ms/step
Epoch 31/500
Epoch 00031: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6837 - val_loss: 0.1006 - lr: 1.0000e-05 - 393ms/epoch - 4ms/step
Epoch 32/500
Epoch 00032: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6834 - val_loss: 0.1007 - lr: 1.0000e-05 - 387ms/epoch - 4ms/step
Epoch 33/500
Epoch 00033: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6830 - val_loss: 0.1009 - lr: 1.0000e-05 - 381ms/epoch - 4ms/step
Epoch 34/500
Epoch 00034: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6827 - val_loss: 0.1011 - lr: 1.0000e-05 - 389ms/epoch - 4ms/step
Epoch 35/500
Epoch 00035: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6823 - val_loss: 0.1013 - lr: 1.0000e-05 - 405ms/epoch - 4ms/step
Epoch 36/500
Epoch 00036: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6820 - val_loss: 0.1015 - lr: 1.0000e-05 - 400ms/epoch - 4ms/step
Epoch 37/500
Epoch 00037: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6816 - val_loss: 0.1017 - lr: 1.0000e-05 - 394ms/epoch - 4ms/step
Epoch 38/500
Epoch 00038: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6812 - val_loss: 0.1019 - lr: 1.0000e-05 - 398ms/epoch - 4ms/step
Epoch 39/500
Epoch 00039: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6809 - val_loss: 0.1021 - lr: 1.0000e-05 - 399ms/epoch - 4ms/step
Epoch 40/500
Epoch 00040: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6805 - val_loss: 0.1023 - lr: 1.0000e-05 - 400ms/epoch - 4ms/step
Epoch 41/500
Epoch 00041: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6802 - val_loss: 0.1026 - lr: 1.0000e-05 - 402ms/epoch - 4ms/step
Epoch 42/500
Epoch 00042: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6798 - val_loss: 0.1028 - lr: 1.0000e-05 - 384ms/epoch - 4ms/step
Epoch 43/500
Epoch 00043: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6794 - val_loss: 0.1030 - lr: 1.0000e-05 - 383ms/epoch - 4ms/step
Epoch 44/500
Epoch 00044: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6791 - val_loss: 0.1032 - lr: 1.0000e-05 - 409ms/epoch - 5ms/step
Epoch 45/500
Epoch 00045: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6787 - val_loss: 0.1034 - lr: 1.0000e-05 - 394ms/epoch - 4ms/step
Epoch 46/500
Epoch 00046: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6784 - val_loss: 0.1036 - lr: 1.0000e-05 - 401ms/epoch - 4ms/step
Epoch 47/500
Epoch 00047: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6780 - val_loss: 0.1039 - lr: 1.0000e-05 - 392ms/epoch - 4ms/step
Epoch 48/500
Epoch 00048: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6777 - val_loss: 0.1041 - lr: 1.0000e-05 - 380ms/epoch - 4ms/step
Epoch 49/500
Epoch 00049: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6773 - val_loss: 0.1043 - lr: 1.0000e-05 - 391ms/epoch - 4ms/step
Epoch 50/500
Epoch 00050: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6770 - val_loss: 0.1045 - lr: 1.0000e-05 - 388ms/epoch - 4ms/step
Epoch 51/500
Epoch 00051: val_loss did not improve from 0.04701
90/90 - 0s - loss: 0.6766 - val_loss: 0.1048 - lr: 1.0000e-05 - 400ms/epoch - 4ms/step
Epoch 00051: early stopping
SMA Prediction vs Close: 54.85% Accuracy Prediction vs Prediction: 47.01% Accuracy MSE: 27.593274833467863 RMSE: 5.252930118844897 MAPE: 4.117405060238624 EMA Prediction vs Close: 54.85% Accuracy Prediction vs Prediction: 49.63% Accuracy MSE: 36.00024912834101 RMSE: 6.000020760659167 MAPE: 4.70426166066095 WMA Prediction vs Close: 54.48% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 41.43257052147296 RMSE: 6.4368136932393005 MAPE: 5.073793611210466 DEMA Prediction vs Close: 52.24% Accuracy Prediction vs Prediction: 45.15% Accuracy MSE: 181.24518187853474 RMSE: 13.462733076108087 MAPE: 12.35792879577932 KAMA Prediction vs Close: 56.72% Accuracy Prediction vs Prediction: 46.27% Accuracy MSE: 27.433326731723586 RMSE: 5.237683336335214 MAPE: 4.154161853007022 MIDPOINT Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 44.78% Accuracy MSE: 17.391761137690516 RMSE: 4.170343047962663 MAPE: 3.393694867236689 T3 Prediction vs Close: 53.73% Accuracy Prediction vs Prediction: 47.76% Accuracy MSE: 65.97403618729489 RMSE: 8.122440285240321 MAPE: 6.528054740759965 TEMA Prediction vs Close: 51.87% Accuracy Prediction vs Prediction: 47.76% Accuracy MSE: 23.78118431614061 RMSE: 4.8765955661855545 MAPE: 4.3174203405459135 Runtime: mins: 41.84255483309998
Architecture Used¶
In [ ]:
from google.colab import files
import cv2
uploaded = files.upload()
In [ ]:
img = cv2.imread('Experiment5.png')
plt.figure(figsize=(20,10))
plt.axis("off")
plt.title('LSTM Architecture '+imgfile,fontsize=18)
plt.imshow(img)
Model Plots¶
In [85]:
with open('simulation8_data.json') as json_file:
simulation8 = json.load(json_file)
fileimg = 'Experiment8'
In [86]:
for i in range(len(list(simulation8.keys()))):
SIM = list(simulation8.keys())[i]
plot_train(simulation8,SIM)
plot_test(simulation8,SIM)
----- Train RMSE for SMA ----- 19.309560563509795 ----- Train_MSE_LSTM for SMA ----- 372.8591291558527 ----- Train MAE LSTM for SMA ----- 19.28235636135139
----- Test RMSE for SMA----- 5.252930118844897 ----- Test_MSE_LSTM for SMA----- 27.593274833467863 ----- Test_MAE_LSTM for SMA----- 4.117405060238624
----- Train RMSE for EMA ----- 23.393549647032383 ----- Train_MSE_LSTM for EMA ----- 547.258165088169 ----- Train MAE LSTM for EMA ----- 23.379056536325134
----- Test RMSE for EMA----- 6.000020760659167 ----- Test_MSE_LSTM for EMA----- 36.00024912834101 ----- Test_MAE_LSTM for EMA----- 4.70426166066095
----- Train RMSE for WMA ----- 21.96266675676863 ----- Train_MSE_LSTM for WMA ----- 482.35873106886993 ----- Train MAE LSTM for WMA ----- 21.93508726535457
----- Test RMSE for WMA----- 6.4368136932393005 ----- Test_MSE_LSTM for WMA----- 41.43257052147296 ----- Test_MAE_LSTM for WMA----- 5.073793611210466
----- Train RMSE for DEMA ----- 25.637680841649903 ----- Train_MSE_LSTM for DEMA ----- 657.2906789383026 ----- Train MAE LSTM for DEMA ----- 25.614362844146125
----- Test RMSE for DEMA----- 13.462733076108087 ----- Test_MSE_LSTM for DEMA----- 181.24518187853474 ----- Test_MAE_LSTM for DEMA----- 12.35792879577932
----- Train RMSE for KAMA ----- 15.645649078988193 ----- Train_MSE_LSTM for KAMA ----- 244.78633510284408 ----- Train MAE LSTM for KAMA ----- 15.537092394167834
----- Test RMSE for KAMA----- 5.237683336335214 ----- Test_MSE_LSTM for KAMA----- 27.433326731723586 ----- Test_MAE_LSTM for KAMA----- 4.154161853007022
----- Train RMSE for MIDPOINT ----- 15.708219768951658 ----- Train_MSE_LSTM for MIDPOINT ----- 246.74816830968368 ----- Train MAE LSTM for MIDPOINT ----- 15.633552206624852
----- Test RMSE for MIDPOINT----- 4.170343047962663 ----- Test_MSE_LSTM for MIDPOINT----- 17.391761137690516 ----- Test_MAE_LSTM for MIDPOINT----- 3.393694867236689
----- Train RMSE for T3 ----- 23.305790930247145 ----- Train_MSE_LSTM for T3 ----- 543.15989088439 ----- Train MAE LSTM for T3 ----- 23.296647801257595
----- Test RMSE for T3----- 8.122440285240321 ----- Test_MSE_LSTM for T3----- 65.97403618729489 ----- Test_MAE_LSTM for T3----- 6.528054740759965
----- Train RMSE for TEMA ----- 19.052881062333018 ----- Train_MSE_LSTM for TEMA ----- 363.0122767754081 ----- Train MAE LSTM for TEMA ----- 19.014995029657193
----- Test RMSE for TEMA----- 4.8765955661855545 ----- Test_MSE_LSTM for TEMA----- 23.78118431614061 ----- Test_MAE_LSTM for TEMA----- 4.3174203405459135
List of RMSE, MSE & MAE scores for Test data¶
In [ ]:
import json
with open('simulation1_data.json') as json_file:
simulation1 = json.load(json_file)
with open('simulation2_data.json') as json_file:
simulation2 = json.load(json_file)
with open('simulation3_data.json') as json_file:
simulation3 = json.load(json_file)
with open('simulation4_data.json') as json_file:
simulation4 = json.load(json_file)
with open('simulation5_data.json') as json_file:
simulation5 = json.load(json_file)
with open('simulation6_data.json') as json_file:
simulation6 = json.load(json_file)
with open('simulation7_data.json') as json_file:
simulation7 = json.load(json_file)
with open('simulation8_data.json') as json_file:
simulation8 = json.load(json_file)
In [ ]:
text = 'Stock with Google Trends '
simulations = [simulation1,simulation2,simulation3,simulation4,simulation5,simulation6,simulation7,simulation8]
for i,simulation in enumerate(simulations):
for ma in simulation.keys():
print(text+'Experiment ',i+1,' for MA :',ma,'the MSE is: ',simulation[ma]['final']['mse'])
print(text+'Experiment ',i+1,' for MA :',ma,'the RMSE is: ',simulation[ma]['final']['rmse'])
print(text+'Experiment ',i+1,' for MA :',ma,'the MAE is: ',simulation[ma]['final']['mae'])
Stock with Google Trends Experiment 1 for MA : SMA the MSE is: 216.26215770312788 Stock with Google Trends Experiment 1 for MA : SMA the RMSE is: 14.705854538350632 Stock with Google Trends Experiment 1 for MA : SMA the MAE is: 11.92367463073216 Stock with Google Trends Experiment 1 for MA : EMA the MSE is: 122.96033735804009 Stock with Google Trends Experiment 1 for MA : EMA the RMSE is: 11.08874823224155 Stock with Google Trends Experiment 1 for MA : EMA the MAE is: 9.251696034357076 Stock with Google Trends Experiment 1 for MA : WMA the MSE is: 56.80196950717294 Stock with Google Trends Experiment 1 for MA : WMA the RMSE is: 7.5367081346681415 Stock with Google Trends Experiment 1 for MA : WMA the MAE is: 5.956122993340066 Stock with Google Trends Experiment 1 for MA : DEMA the MSE is: 120.51656357671082 Stock with Google Trends Experiment 1 for MA : DEMA the RMSE is: 10.978003624371365 Stock with Google Trends Experiment 1 for MA : DEMA the MAE is: 9.343426819843298 Stock with Google Trends Experiment 1 for MA : KAMA the MSE is: 80.73205723036311 Stock with Google Trends Experiment 1 for MA : KAMA the RMSE is: 8.985101959931402 Stock with Google Trends Experiment 1 for MA : KAMA the MAE is: 7.079003376879244 Stock with Google Trends Experiment 1 for MA : MIDPOINT the MSE is: 75.80130571921515 Stock with Google Trends Experiment 1 for MA : MIDPOINT the RMSE is: 8.70639453041356 Stock with Google Trends Experiment 1 for MA : MIDPOINT the MAE is: 7.130945881105426 Stock with Google Trends Experiment 1 for MA : T3 the MSE is: 46.36148283886003 Stock with Google Trends Experiment 1 for MA : T3 the RMSE is: 6.808926702415001 Stock with Google Trends Experiment 1 for MA : T3 the MAE is: 5.596706034896943 Stock with Google Trends Experiment 1 for MA : TEMA the MSE is: 32.02006831549358 Stock with Google Trends Experiment 1 for MA : TEMA the RMSE is: 5.658627776722337 Stock with Google Trends Experiment 1 for MA : TEMA the MAE is: 4.840005256997922 Stock with Google Trends Experiment 2 for MA : SMA the MSE is: 136.8787627980392 Stock with Google Trends Experiment 2 for MA : SMA the RMSE is: 11.699519767838302 Stock with Google Trends Experiment 2 for MA : SMA the MAE is: 9.782684450137419 Stock with Google Trends Experiment 2 for MA : EMA the MSE is: 325.2876788257106 Stock with Google Trends Experiment 2 for MA : EMA the RMSE is: 18.035733387520192 Stock with Google Trends Experiment 2 for MA : EMA the MAE is: 15.58326491461623 Stock with Google Trends Experiment 2 for MA : WMA the MSE is: 142.26654445774975 Stock with Google Trends Experiment 2 for MA : WMA the RMSE is: 11.927554001460221 Stock with Google Trends Experiment 2 for MA : WMA the MAE is: 9.787176512242624 Stock with Google Trends Experiment 2 for MA : DEMA the MSE is: 171.26734938505615 Stock with Google Trends Experiment 2 for MA : DEMA the RMSE is: 13.086915197442679 Stock with Google Trends Experiment 2 for MA : DEMA the MAE is: 11.821213958102536 Stock with Google Trends Experiment 2 for MA : KAMA the MSE is: 52.24625253706662 Stock with Google Trends Experiment 2 for MA : KAMA the RMSE is: 7.228156925321048 Stock with Google Trends Experiment 2 for MA : KAMA the MAE is: 5.787764162926424 Stock with Google Trends Experiment 2 for MA : MIDPOINT the MSE is: 51.03620830548878 Stock with Google Trends Experiment 2 for MA : MIDPOINT the RMSE is: 7.143963067197981 Stock with Google Trends Experiment 2 for MA : MIDPOINT the MAE is: 5.7689786092745114 Stock with Google Trends Experiment 2 for MA : T3 the MSE is: 131.82692716984482 Stock with Google Trends Experiment 2 for MA : T3 the RMSE is: 11.481590794391028 Stock with Google Trends Experiment 2 for MA : T3 the MAE is: 9.148826908925223 Stock with Google Trends Experiment 2 for MA : TEMA the MSE is: 103.01748696218638 Stock with Google Trends Experiment 2 for MA : TEMA the RMSE is: 10.149753049320283 Stock with Google Trends Experiment 2 for MA : TEMA the MAE is: 9.012271728014667 Stock with Google Trends Experiment 3 for MA : SMA the MSE is: 60.72793966434877 Stock with Google Trends Experiment 3 for MA : SMA the RMSE is: 7.792813334370892 Stock with Google Trends Experiment 3 for MA : SMA the MAE is: 6.245110563496496 Stock with Google Trends Experiment 3 for MA : EMA the MSE is: 24.31424046870049 Stock with Google Trends Experiment 3 for MA : EMA the RMSE is: 4.930947218202654 Stock with Google Trends Experiment 3 for MA : EMA the MAE is: 4.072073008160127 Stock with Google Trends Experiment 3 for MA : WMA the MSE is: 68.15170187847146 Stock with Google Trends Experiment 3 for MA : WMA the RMSE is: 8.255404404296101 Stock with Google Trends Experiment 3 for MA : WMA the MAE is: 6.806281257852447 Stock with Google Trends Experiment 3 for MA : DEMA the MSE is: 43.922177447330554 Stock with Google Trends Experiment 3 for MA : DEMA the RMSE is: 6.627380888958364 Stock with Google Trends Experiment 3 for MA : DEMA the MAE is: 5.414540694927783 Stock with Google Trends Experiment 3 for MA : KAMA the MSE is: 23.99643218488633 Stock with Google Trends Experiment 3 for MA : KAMA the RMSE is: 4.8986153334270215 Stock with Google Trends Experiment 3 for MA : KAMA the MAE is: 3.8674202618000764 Stock with Google Trends Experiment 3 for MA : MIDPOINT the MSE is: 28.7191654730941 Stock with Google Trends Experiment 3 for MA : MIDPOINT the RMSE is: 5.359026541555295 Stock with Google Trends Experiment 3 for MA : MIDPOINT the MAE is: 4.42030651732032 Stock with Google Trends Experiment 3 for MA : T3 the MSE is: 57.05690799124099 Stock with Google Trends Experiment 3 for MA : T3 the RMSE is: 7.553602318843705 Stock with Google Trends Experiment 3 for MA : T3 the MAE is: 6.064989593585796 Stock with Google Trends Experiment 3 for MA : TEMA the MSE is: 20.964688340236016 Stock with Google Trends Experiment 3 for MA : TEMA the RMSE is: 4.578721256009806 Stock with Google Trends Experiment 3 for MA : TEMA the MAE is: 3.7212897315589664 Stock with Google Trends Experiment 4 for MA : SMA the MSE is: 23.31830336349202 Stock with Google Trends Experiment 4 for MA : SMA the RMSE is: 4.828902915103183 Stock with Google Trends Experiment 4 for MA : SMA the MAE is: 3.806885992834059 Stock with Google Trends Experiment 4 for MA : EMA the MSE is: 31.4391560756623 Stock with Google Trends Experiment 4 for MA : EMA the RMSE is: 5.607063052584865 Stock with Google Trends Experiment 4 for MA : EMA the MAE is: 4.398444723456604 Stock with Google Trends Experiment 4 for MA : WMA the MSE is: 48.85272948439791 Stock with Google Trends Experiment 4 for MA : WMA the RMSE is: 6.989472761546318 Stock with Google Trends Experiment 4 for MA : WMA the MAE is: 5.616901258925532 Stock with Google Trends Experiment 4 for MA : DEMA the MSE is: 143.4471215002686 Stock with Google Trends Experiment 4 for MA : DEMA the RMSE is: 11.976941241413376 Stock with Google Trends Experiment 4 for MA : DEMA the MAE is: 10.686872819228396 Stock with Google Trends Experiment 4 for MA : KAMA the MSE is: 23.251670970583447 Stock with Google Trends Experiment 4 for MA : KAMA the RMSE is: 4.821998648961181 Stock with Google Trends Experiment 4 for MA : KAMA the MAE is: 3.833042253232743 Stock with Google Trends Experiment 4 for MA : MIDPOINT the MSE is: 16.39872837560197 Stock with Google Trends Experiment 4 for MA : MIDPOINT the RMSE is: 4.0495343405880595 Stock with Google Trends Experiment 4 for MA : MIDPOINT the MAE is: 3.299619771312048 Stock with Google Trends Experiment 4 for MA : T3 the MSE is: 85.43536380908036 Stock with Google Trends Experiment 4 for MA : T3 the RMSE is: 9.24312521872772 Stock with Google Trends Experiment 4 for MA : T3 the MAE is: 7.5496901284439915 Stock with Google Trends Experiment 4 for MA : TEMA the MSE is: 17.749245841001986 Stock with Google Trends Experiment 4 for MA : TEMA the RMSE is: 4.21298538343085 Stock with Google Trends Experiment 4 for MA : TEMA the MAE is: 3.636908590169574 Stock with Google Trends Experiment 5 for MA : SMA the MSE is: 38.984836670221576 Stock with Google Trends Experiment 5 for MA : SMA the RMSE is: 6.24378384236847 Stock with Google Trends Experiment 5 for MA : SMA the MAE is: 5.10393861237253 Stock with Google Trends Experiment 5 for MA : EMA the MSE is: 124.25764226707467 Stock with Google Trends Experiment 5 for MA : EMA the RMSE is: 11.1470912020614 Stock with Google Trends Experiment 5 for MA : EMA the MAE is: 9.17724208981177 Stock with Google Trends Experiment 5 for MA : WMA the MSE is: 38.21405797999008 Stock with Google Trends Experiment 5 for MA : WMA the RMSE is: 6.1817520154071275 Stock with Google Trends Experiment 5 for MA : WMA the MAE is: 5.0592557753421294 Stock with Google Trends Experiment 5 for MA : DEMA the MSE is: 272.80520892035037 Stock with Google Trends Experiment 5 for MA : DEMA the RMSE is: 16.51681594376926 Stock with Google Trends Experiment 5 for MA : DEMA the MAE is: 15.690440427295842 Stock with Google Trends Experiment 5 for MA : KAMA the MSE is: 39.855043502438484 Stock with Google Trends Experiment 5 for MA : KAMA the RMSE is: 6.313085101789654 Stock with Google Trends Experiment 5 for MA : KAMA the MAE is: 4.932118299391016 Stock with Google Trends Experiment 5 for MA : MIDPOINT the MSE is: 247.8059617737088 Stock with Google Trends Experiment 5 for MA : MIDPOINT the RMSE is: 15.741853822650901 Stock with Google Trends Experiment 5 for MA : MIDPOINT the MAE is: 13.137429929578502 Stock with Google Trends Experiment 5 for MA : T3 the MSE is: 210.84512917819418 Stock with Google Trends Experiment 5 for MA : T3 the RMSE is: 14.520507194247527 Stock with Google Trends Experiment 5 for MA : T3 the MAE is: 11.877711162377306 Stock with Google Trends Experiment 5 for MA : TEMA the MSE is: 47.81525777472662 Stock with Google Trends Experiment 5 for MA : TEMA the RMSE is: 6.9148577552055706 Stock with Google Trends Experiment 5 for MA : TEMA the MAE is: 5.805355375153735 Stock with Google Trends Experiment 6 for MA : SMA the MSE is: 71.66891031373025 Stock with Google Trends Experiment 6 for MA : SMA the RMSE is: 8.465749247038342 Stock with Google Trends Experiment 6 for MA : SMA the MAE is: 6.880610177712922 Stock with Google Trends Experiment 6 for MA : EMA the MSE is: 67.24711347278334 Stock with Google Trends Experiment 6 for MA : EMA the RMSE is: 8.200433736869249 Stock with Google Trends Experiment 6 for MA : EMA the MAE is: 6.781803215137433 Stock with Google Trends Experiment 6 for MA : WMA the MSE is: 58.550384022767716 Stock with Google Trends Experiment 6 for MA : WMA the RMSE is: 7.6518222681115455 Stock with Google Trends Experiment 6 for MA : WMA the MAE is: 6.1413991074844 Stock with Google Trends Experiment 6 for MA : DEMA the MSE is: 127.72881036471918 Stock with Google Trends Experiment 6 for MA : DEMA the RMSE is: 11.301717142307146 Stock with Google Trends Experiment 6 for MA : DEMA the MAE is: 10.306940424406019 Stock with Google Trends Experiment 6 for MA : KAMA the MSE is: 43.855468679122 Stock with Google Trends Experiment 6 for MA : KAMA the RMSE is: 6.622346161227303 Stock with Google Trends Experiment 6 for MA : KAMA the MAE is: 5.4751276749367985 Stock with Google Trends Experiment 6 for MA : MIDPOINT the MSE is: 67.44416314351042 Stock with Google Trends Experiment 6 for MA : MIDPOINT the RMSE is: 8.212439536673035 Stock with Google Trends Experiment 6 for MA : MIDPOINT the MAE is: 6.768235104271493 Stock with Google Trends Experiment 6 for MA : T3 the MSE is: 154.873959597027 Stock with Google Trends Experiment 6 for MA : T3 the RMSE is: 12.444836664136135 Stock with Google Trends Experiment 6 for MA : T3 the MAE is: 10.329006454112236 Stock with Google Trends Experiment 6 for MA : TEMA the MSE is: 163.5522149093863 Stock with Google Trends Experiment 6 for MA : TEMA the RMSE is: 12.788753454085597 Stock with Google Trends Experiment 6 for MA : TEMA the MAE is: 11.455728191899736 Stock with Google Trends Experiment 7 for MA : SMA the MSE is: 37.148636819682245 Stock with Google Trends Experiment 7 for MA : SMA the RMSE is: 6.094968155756209 Stock with Google Trends Experiment 7 for MA : SMA the MAE is: 5.090179331518223 Stock with Google Trends Experiment 7 for MA : EMA the MSE is: 31.582654654902484 Stock with Google Trends Experiment 7 for MA : EMA the RMSE is: 5.619844718041815 Stock with Google Trends Experiment 7 for MA : EMA the MAE is: 4.507182634072088 Stock with Google Trends Experiment 7 for MA : WMA the MSE is: 65.00101872981564 Stock with Google Trends Experiment 7 for MA : WMA the RMSE is: 8.062320926992156 Stock with Google Trends Experiment 7 for MA : WMA the MAE is: 6.705711592581163 Stock with Google Trends Experiment 7 for MA : DEMA the MSE is: 35.269002386685244 Stock with Google Trends Experiment 7 for MA : DEMA the RMSE is: 5.938771117553298 Stock with Google Trends Experiment 7 for MA : DEMA the MAE is: 4.62878838931535 Stock with Google Trends Experiment 7 for MA : KAMA the MSE is: 62.25156682504816 Stock with Google Trends Experiment 7 for MA : KAMA the RMSE is: 7.8899662119078915 Stock with Google Trends Experiment 7 for MA : KAMA the MAE is: 6.222956717810362 Stock with Google Trends Experiment 7 for MA : MIDPOINT the MSE is: 84.47531990876952 Stock with Google Trends Experiment 7 for MA : MIDPOINT the RMSE is: 9.191045637399997 Stock with Google Trends Experiment 7 for MA : MIDPOINT the MAE is: 7.890202393641488 Stock with Google Trends Experiment 7 for MA : T3 the MSE is: 92.6070950307407 Stock with Google Trends Experiment 7 for MA : T3 the RMSE is: 9.623258025780078 Stock with Google Trends Experiment 7 for MA : T3 the MAE is: 8.212466968891306 Stock with Google Trends Experiment 7 for MA : TEMA the MSE is: 50.229608880159034 Stock with Google Trends Experiment 7 for MA : TEMA the RMSE is: 7.087285014740061 Stock with Google Trends Experiment 7 for MA : TEMA the MAE is: 6.2877236827531835 Stock with Google Trends Experiment 8 for MA : SMA the MSE is: 27.593274833467863 Stock with Google Trends Experiment 8 for MA : SMA the RMSE is: 5.252930118844897 Stock with Google Trends Experiment 8 for MA : SMA the MAE is: 4.117405060238624 Stock with Google Trends Experiment 8 for MA : EMA the MSE is: 36.00024912834101 Stock with Google Trends Experiment 8 for MA : EMA the RMSE is: 6.000020760659167 Stock with Google Trends Experiment 8 for MA : EMA the MAE is: 4.70426166066095 Stock with Google Trends Experiment 8 for MA : WMA the MSE is: 41.43257052147296 Stock with Google Trends Experiment 8 for MA : WMA the RMSE is: 6.4368136932393005 Stock with Google Trends Experiment 8 for MA : WMA the MAE is: 5.073793611210466 Stock with Google Trends Experiment 8 for MA : DEMA the MSE is: 181.24518187853474 Stock with Google Trends Experiment 8 for MA : DEMA the RMSE is: 13.462733076108087 Stock with Google Trends Experiment 8 for MA : DEMA the MAE is: 12.35792879577932 Stock with Google Trends Experiment 8 for MA : KAMA the MSE is: 27.433326731723586 Stock with Google Trends Experiment 8 for MA : KAMA the RMSE is: 5.237683336335214 Stock with Google Trends Experiment 8 for MA : KAMA the MAE is: 4.154161853007022 Stock with Google Trends Experiment 8 for MA : MIDPOINT the MSE is: 17.391761137690516 Stock with Google Trends Experiment 8 for MA : MIDPOINT the RMSE is: 4.170343047962663 Stock with Google Trends Experiment 8 for MA : MIDPOINT the MAE is: 3.393694867236689 Stock with Google Trends Experiment 8 for MA : T3 the MSE is: 65.97403618729489 Stock with Google Trends Experiment 8 for MA : T3 the RMSE is: 8.122440285240321 Stock with Google Trends Experiment 8 for MA : T3 the MAE is: 6.528054740759965 Stock with Google Trends Experiment 8 for MA : TEMA the MSE is: 23.78118431614061 Stock with Google Trends Experiment 8 for MA : TEMA the RMSE is: 4.8765955661855545 Stock with Google Trends Experiment 8 for MA : TEMA the MAE is: 4.3174203405459135
In [ ]:
text = 'Stock with Google Trends '
simulations = [simulation1,simulation2,simulation3,simulation4,simulation5,simulation6,simulation7,simulation8]
for i,simulation in enumerate(simulations):
for ma in simulation.keys():
# print(text+'Experiment ',i+1,' for MA :',ma,'the MSE is: ',simulation[ma]['final']['mse'])
print(text+'Experiment ',i+1,' for MA :',ma,'the RMSE is: ',simulation[ma]['final']['rmse'])
# print(text+'Experiment ',i+1,' for MA :',ma,'the MAE is: ',simulation[ma]['final']['mae'])
for ma in simulation.keys():
print(text+'Experiment ',i+1,' for MA :',ma,'the MSE is: ',simulation[ma]['final']['mse'])
# print(text+'Experiment ',i+1,' for MA :',ma,'the RMSE is: ',simulation[ma]['final']['rmse'])
# print(text+'Experiment ',i+1,' for MA :',ma,'the MAE is: ',simulation[ma]['final']['mae'])
for ma in simulation.keys():
# print(text+'Experiment ',i+1,' for MA :',ma,'the MSE is: ',simulation[ma]['final']['mse'])
# print(text+'Experiment ',i+1,' for MA :',ma,'the RMSE is: ',simulation[ma]['final']['rmse'])
print(text+'Experiment ',i+1,' for MA :',ma,'the MAE is: ',simulation[ma]['final']['mae'])
Stock with Google Trends Experiment 1 for MA : SMA the RMSE is: 14.705854538350632 Stock with Google Trends Experiment 1 for MA : EMA the RMSE is: 11.08874823224155 Stock with Google Trends Experiment 1 for MA : WMA the RMSE is: 7.5367081346681415 Stock with Google Trends Experiment 1 for MA : DEMA the RMSE is: 10.978003624371365 Stock with Google Trends Experiment 1 for MA : KAMA the RMSE is: 8.985101959931402 Stock with Google Trends Experiment 1 for MA : MIDPOINT the RMSE is: 8.70639453041356 Stock with Google Trends Experiment 1 for MA : T3 the RMSE is: 6.808926702415001 Stock with Google Trends Experiment 1 for MA : TEMA the RMSE is: 5.658627776722337 Stock with Google Trends Experiment 1 for MA : SMA the MSE is: 216.26215770312788 Stock with Google Trends Experiment 1 for MA : EMA the MSE is: 122.96033735804009 Stock with Google Trends Experiment 1 for MA : WMA the MSE is: 56.80196950717294 Stock with Google Trends Experiment 1 for MA : DEMA the MSE is: 120.51656357671082 Stock with Google Trends Experiment 1 for MA : KAMA the MSE is: 80.73205723036311 Stock with Google Trends Experiment 1 for MA : MIDPOINT the MSE is: 75.80130571921515 Stock with Google Trends Experiment 1 for MA : T3 the MSE is: 46.36148283886003 Stock with Google Trends Experiment 1 for MA : TEMA the MSE is: 32.02006831549358 Stock with Google Trends Experiment 1 for MA : SMA the MAE is: 11.92367463073216 Stock with Google Trends Experiment 1 for MA : EMA the MAE is: 9.251696034357076 Stock with Google Trends Experiment 1 for MA : WMA the MAE is: 5.956122993340066 Stock with Google Trends Experiment 1 for MA : DEMA the MAE is: 9.343426819843298 Stock with Google Trends Experiment 1 for MA : KAMA the MAE is: 7.079003376879244 Stock with Google Trends Experiment 1 for MA : MIDPOINT the MAE is: 7.130945881105426 Stock with Google Trends Experiment 1 for MA : T3 the MAE is: 5.596706034896943 Stock with Google Trends Experiment 1 for MA : TEMA the MAE is: 4.840005256997922 Stock with Google Trends Experiment 2 for MA : SMA the RMSE is: 11.699519767838302 Stock with Google Trends Experiment 2 for MA : EMA the RMSE is: 18.035733387520192 Stock with Google Trends Experiment 2 for MA : WMA the RMSE is: 11.927554001460221 Stock with Google Trends Experiment 2 for MA : DEMA the RMSE is: 13.086915197442679 Stock with Google Trends Experiment 2 for MA : KAMA the RMSE is: 7.228156925321048 Stock with Google Trends Experiment 2 for MA : MIDPOINT the RMSE is: 7.143963067197981 Stock with Google Trends Experiment 2 for MA : T3 the RMSE is: 11.481590794391028 Stock with Google Trends Experiment 2 for MA : TEMA the RMSE is: 10.149753049320283 Stock with Google Trends Experiment 2 for MA : SMA the MSE is: 136.8787627980392 Stock with Google Trends Experiment 2 for MA : EMA the MSE is: 325.2876788257106 Stock with Google Trends Experiment 2 for MA : WMA the MSE is: 142.26654445774975 Stock with Google Trends Experiment 2 for MA : DEMA the MSE is: 171.26734938505615 Stock with Google Trends Experiment 2 for MA : KAMA the MSE is: 52.24625253706662 Stock with Google Trends Experiment 2 for MA : MIDPOINT the MSE is: 51.03620830548878 Stock with Google Trends Experiment 2 for MA : T3 the MSE is: 131.82692716984482 Stock with Google Trends Experiment 2 for MA : TEMA the MSE is: 103.01748696218638 Stock with Google Trends Experiment 2 for MA : SMA the MAE is: 9.782684450137419 Stock with Google Trends Experiment 2 for MA : EMA the MAE is: 15.58326491461623 Stock with Google Trends Experiment 2 for MA : WMA the MAE is: 9.787176512242624 Stock with Google Trends Experiment 2 for MA : DEMA the MAE is: 11.821213958102536 Stock with Google Trends Experiment 2 for MA : KAMA the MAE is: 5.787764162926424 Stock with Google Trends Experiment 2 for MA : MIDPOINT the MAE is: 5.7689786092745114 Stock with Google Trends Experiment 2 for MA : T3 the MAE is: 9.148826908925223 Stock with Google Trends Experiment 2 for MA : TEMA the MAE is: 9.012271728014667 Stock with Google Trends Experiment 3 for MA : SMA the RMSE is: 7.792813334370892 Stock with Google Trends Experiment 3 for MA : EMA the RMSE is: 4.930947218202654 Stock with Google Trends Experiment 3 for MA : WMA the RMSE is: 8.255404404296101 Stock with Google Trends Experiment 3 for MA : DEMA the RMSE is: 6.627380888958364 Stock with Google Trends Experiment 3 for MA : KAMA the RMSE is: 4.8986153334270215 Stock with Google Trends Experiment 3 for MA : MIDPOINT the RMSE is: 5.359026541555295 Stock with Google Trends Experiment 3 for MA : T3 the RMSE is: 7.553602318843705 Stock with Google Trends Experiment 3 for MA : TEMA the RMSE is: 4.578721256009806 Stock with Google Trends Experiment 3 for MA : SMA the MSE is: 60.72793966434877 Stock with Google Trends Experiment 3 for MA : EMA the MSE is: 24.31424046870049 Stock with Google Trends Experiment 3 for MA : WMA the MSE is: 68.15170187847146 Stock with Google Trends Experiment 3 for MA : DEMA the MSE is: 43.922177447330554 Stock with Google Trends Experiment 3 for MA : KAMA the MSE is: 23.99643218488633 Stock with Google Trends Experiment 3 for MA : MIDPOINT the MSE is: 28.7191654730941 Stock with Google Trends Experiment 3 for MA : T3 the MSE is: 57.05690799124099 Stock with Google Trends Experiment 3 for MA : TEMA the MSE is: 20.964688340236016 Stock with Google Trends Experiment 3 for MA : SMA the MAE is: 6.245110563496496 Stock with Google Trends Experiment 3 for MA : EMA the MAE is: 4.072073008160127 Stock with Google Trends Experiment 3 for MA : WMA the MAE is: 6.806281257852447 Stock with Google Trends Experiment 3 for MA : DEMA the MAE is: 5.414540694927783 Stock with Google Trends Experiment 3 for MA : KAMA the MAE is: 3.8674202618000764 Stock with Google Trends Experiment 3 for MA : MIDPOINT the MAE is: 4.42030651732032 Stock with Google Trends Experiment 3 for MA : T3 the MAE is: 6.064989593585796 Stock with Google Trends Experiment 3 for MA : TEMA the MAE is: 3.7212897315589664 Stock with Google Trends Experiment 4 for MA : SMA the RMSE is: 4.828902915103183 Stock with Google Trends Experiment 4 for MA : EMA the RMSE is: 5.607063052584865 Stock with Google Trends Experiment 4 for MA : WMA the RMSE is: 6.989472761546318 Stock with Google Trends Experiment 4 for MA : DEMA the RMSE is: 11.976941241413376 Stock with Google Trends Experiment 4 for MA : KAMA the RMSE is: 4.821998648961181 Stock with Google Trends Experiment 4 for MA : MIDPOINT the RMSE is: 4.0495343405880595 Stock with Google Trends Experiment 4 for MA : T3 the RMSE is: 9.24312521872772 Stock with Google Trends Experiment 4 for MA : TEMA the RMSE is: 4.21298538343085 Stock with Google Trends Experiment 4 for MA : SMA the MSE is: 23.31830336349202 Stock with Google Trends Experiment 4 for MA : EMA the MSE is: 31.4391560756623 Stock with Google Trends Experiment 4 for MA : WMA the MSE is: 48.85272948439791 Stock with Google Trends Experiment 4 for MA : DEMA the MSE is: 143.4471215002686 Stock with Google Trends Experiment 4 for MA : KAMA the MSE is: 23.251670970583447 Stock with Google Trends Experiment 4 for MA : MIDPOINT the MSE is: 16.39872837560197 Stock with Google Trends Experiment 4 for MA : T3 the MSE is: 85.43536380908036 Stock with Google Trends Experiment 4 for MA : TEMA the MSE is: 17.749245841001986 Stock with Google Trends Experiment 4 for MA : SMA the MAE is: 3.806885992834059 Stock with Google Trends Experiment 4 for MA : EMA the MAE is: 4.398444723456604 Stock with Google Trends Experiment 4 for MA : WMA the MAE is: 5.616901258925532 Stock with Google Trends Experiment 4 for MA : DEMA the MAE is: 10.686872819228396 Stock with Google Trends Experiment 4 for MA : KAMA the MAE is: 3.833042253232743 Stock with Google Trends Experiment 4 for MA : MIDPOINT the MAE is: 3.299619771312048 Stock with Google Trends Experiment 4 for MA : T3 the MAE is: 7.5496901284439915 Stock with Google Trends Experiment 4 for MA : TEMA the MAE is: 3.636908590169574 Stock with Google Trends Experiment 5 for MA : SMA the RMSE is: 6.24378384236847 Stock with Google Trends Experiment 5 for MA : EMA the RMSE is: 11.1470912020614 Stock with Google Trends Experiment 5 for MA : WMA the RMSE is: 6.1817520154071275 Stock with Google Trends Experiment 5 for MA : DEMA the RMSE is: 16.51681594376926 Stock with Google Trends Experiment 5 for MA : KAMA the RMSE is: 6.313085101789654 Stock with Google Trends Experiment 5 for MA : MIDPOINT the RMSE is: 15.741853822650901 Stock with Google Trends Experiment 5 for MA : T3 the RMSE is: 14.520507194247527 Stock with Google Trends Experiment 5 for MA : TEMA the RMSE is: 6.9148577552055706 Stock with Google Trends Experiment 5 for MA : SMA the MSE is: 38.984836670221576 Stock with Google Trends Experiment 5 for MA : EMA the MSE is: 124.25764226707467 Stock with Google Trends Experiment 5 for MA : WMA the MSE is: 38.21405797999008 Stock with Google Trends Experiment 5 for MA : DEMA the MSE is: 272.80520892035037 Stock with Google Trends Experiment 5 for MA : KAMA the MSE is: 39.855043502438484 Stock with Google Trends Experiment 5 for MA : MIDPOINT the MSE is: 247.8059617737088 Stock with Google Trends Experiment 5 for MA : T3 the MSE is: 210.84512917819418 Stock with Google Trends Experiment 5 for MA : TEMA the MSE is: 47.81525777472662 Stock with Google Trends Experiment 5 for MA : SMA the MAE is: 5.10393861237253 Stock with Google Trends Experiment 5 for MA : EMA the MAE is: 9.17724208981177 Stock with Google Trends Experiment 5 for MA : WMA the MAE is: 5.0592557753421294 Stock with Google Trends Experiment 5 for MA : DEMA the MAE is: 15.690440427295842 Stock with Google Trends Experiment 5 for MA : KAMA the MAE is: 4.932118299391016 Stock with Google Trends Experiment 5 for MA : MIDPOINT the MAE is: 13.137429929578502 Stock with Google Trends Experiment 5 for MA : T3 the MAE is: 11.877711162377306 Stock with Google Trends Experiment 5 for MA : TEMA the MAE is: 5.805355375153735 Stock with Google Trends Experiment 6 for MA : SMA the RMSE is: 8.465749247038342 Stock with Google Trends Experiment 6 for MA : EMA the RMSE is: 8.200433736869249 Stock with Google Trends Experiment 6 for MA : WMA the RMSE is: 7.6518222681115455 Stock with Google Trends Experiment 6 for MA : DEMA the RMSE is: 11.301717142307146 Stock with Google Trends Experiment 6 for MA : KAMA the RMSE is: 6.622346161227303 Stock with Google Trends Experiment 6 for MA : MIDPOINT the RMSE is: 8.212439536673035 Stock with Google Trends Experiment 6 for MA : T3 the RMSE is: 12.444836664136135 Stock with Google Trends Experiment 6 for MA : TEMA the RMSE is: 12.788753454085597 Stock with Google Trends Experiment 6 for MA : SMA the MSE is: 71.66891031373025 Stock with Google Trends Experiment 6 for MA : EMA the MSE is: 67.24711347278334 Stock with Google Trends Experiment 6 for MA : WMA the MSE is: 58.550384022767716 Stock with Google Trends Experiment 6 for MA : DEMA the MSE is: 127.72881036471918 Stock with Google Trends Experiment 6 for MA : KAMA the MSE is: 43.855468679122 Stock with Google Trends Experiment 6 for MA : MIDPOINT the MSE is: 67.44416314351042 Stock with Google Trends Experiment 6 for MA : T3 the MSE is: 154.873959597027 Stock with Google Trends Experiment 6 for MA : TEMA the MSE is: 163.5522149093863 Stock with Google Trends Experiment 6 for MA : SMA the MAE is: 6.880610177712922 Stock with Google Trends Experiment 6 for MA : EMA the MAE is: 6.781803215137433 Stock with Google Trends Experiment 6 for MA : WMA the MAE is: 6.1413991074844 Stock with Google Trends Experiment 6 for MA : DEMA the MAE is: 10.306940424406019 Stock with Google Trends Experiment 6 for MA : KAMA the MAE is: 5.4751276749367985 Stock with Google Trends Experiment 6 for MA : MIDPOINT the MAE is: 6.768235104271493 Stock with Google Trends Experiment 6 for MA : T3 the MAE is: 10.329006454112236 Stock with Google Trends Experiment 6 for MA : TEMA the MAE is: 11.455728191899736 Stock with Google Trends Experiment 7 for MA : SMA the RMSE is: 6.094968155756209 Stock with Google Trends Experiment 7 for MA : EMA the RMSE is: 5.619844718041815 Stock with Google Trends Experiment 7 for MA : WMA the RMSE is: 8.062320926992156 Stock with Google Trends Experiment 7 for MA : DEMA the RMSE is: 5.938771117553298 Stock with Google Trends Experiment 7 for MA : KAMA the RMSE is: 7.8899662119078915 Stock with Google Trends Experiment 7 for MA : MIDPOINT the RMSE is: 9.191045637399997 Stock with Google Trends Experiment 7 for MA : T3 the RMSE is: 9.623258025780078 Stock with Google Trends Experiment 7 for MA : TEMA the RMSE is: 7.087285014740061 Stock with Google Trends Experiment 7 for MA : SMA the MSE is: 37.148636819682245 Stock with Google Trends Experiment 7 for MA : EMA the MSE is: 31.582654654902484 Stock with Google Trends Experiment 7 for MA : WMA the MSE is: 65.00101872981564 Stock with Google Trends Experiment 7 for MA : DEMA the MSE is: 35.269002386685244 Stock with Google Trends Experiment 7 for MA : KAMA the MSE is: 62.25156682504816 Stock with Google Trends Experiment 7 for MA : MIDPOINT the MSE is: 84.47531990876952 Stock with Google Trends Experiment 7 for MA : T3 the MSE is: 92.6070950307407 Stock with Google Trends Experiment 7 for MA : TEMA the MSE is: 50.229608880159034 Stock with Google Trends Experiment 7 for MA : SMA the MAE is: 5.090179331518223 Stock with Google Trends Experiment 7 for MA : EMA the MAE is: 4.507182634072088 Stock with Google Trends Experiment 7 for MA : WMA the MAE is: 6.705711592581163 Stock with Google Trends Experiment 7 for MA : DEMA the MAE is: 4.62878838931535 Stock with Google Trends Experiment 7 for MA : KAMA the MAE is: 6.222956717810362 Stock with Google Trends Experiment 7 for MA : MIDPOINT the MAE is: 7.890202393641488 Stock with Google Trends Experiment 7 for MA : T3 the MAE is: 8.212466968891306 Stock with Google Trends Experiment 7 for MA : TEMA the MAE is: 6.2877236827531835 Stock with Google Trends Experiment 8 for MA : SMA the RMSE is: 5.252930118844897 Stock with Google Trends Experiment 8 for MA : EMA the RMSE is: 6.000020760659167 Stock with Google Trends Experiment 8 for MA : WMA the RMSE is: 6.4368136932393005 Stock with Google Trends Experiment 8 for MA : DEMA the RMSE is: 13.462733076108087 Stock with Google Trends Experiment 8 for MA : KAMA the RMSE is: 5.237683336335214 Stock with Google Trends Experiment 8 for MA : MIDPOINT the RMSE is: 4.170343047962663 Stock with Google Trends Experiment 8 for MA : T3 the RMSE is: 8.122440285240321 Stock with Google Trends Experiment 8 for MA : TEMA the RMSE is: 4.8765955661855545 Stock with Google Trends Experiment 8 for MA : SMA the MSE is: 27.593274833467863 Stock with Google Trends Experiment 8 for MA : EMA the MSE is: 36.00024912834101 Stock with Google Trends Experiment 8 for MA : WMA the MSE is: 41.43257052147296 Stock with Google Trends Experiment 8 for MA : DEMA the MSE is: 181.24518187853474 Stock with Google Trends Experiment 8 for MA : KAMA the MSE is: 27.433326731723586 Stock with Google Trends Experiment 8 for MA : MIDPOINT the MSE is: 17.391761137690516 Stock with Google Trends Experiment 8 for MA : T3 the MSE is: 65.97403618729489 Stock with Google Trends Experiment 8 for MA : TEMA the MSE is: 23.78118431614061 Stock with Google Trends Experiment 8 for MA : SMA the MAE is: 4.117405060238624 Stock with Google Trends Experiment 8 for MA : EMA the MAE is: 4.70426166066095 Stock with Google Trends Experiment 8 for MA : WMA the MAE is: 5.073793611210466 Stock with Google Trends Experiment 8 for MA : DEMA the MAE is: 12.35792879577932 Stock with Google Trends Experiment 8 for MA : KAMA the MAE is: 4.154161853007022 Stock with Google Trends Experiment 8 for MA : MIDPOINT the MAE is: 3.393694867236689 Stock with Google Trends Experiment 8 for MA : T3 the MAE is: 6.528054740759965 Stock with Google Trends Experiment 8 for MA : TEMA the MAE is: 4.3174203405459135
Create HTML¶
In [90]:
cd ..
/content/drive/.shortcut-targets-by-id/1IaGjVBlTspxI2CHSrxfYnaiYvsaG0pHs/Stock price prediction/Archana - LSTM Hybrid
In [ ]:
cd drive/MyDrive/Stock price prediction/Archana - LSTM Hybrid
In [93]:
%%shell
jupyter nbconvert --to html LSTM_Hybrid_using_TA_LIB_Google_Trends.ipynb
[NbConvertApp] Converting notebook LSTM_Hybrid_using_TA_LIB_Google_Trends.ipynb to html
[NbConvertApp] ERROR | Notebook JSON is invalid: Additional properties are not allowed (u'metadata' was unexpected)
Failed validating u'additionalProperties' in stream:
On instance[u'cells'][155][u'outputs'][0]:
{u'metadata': {u'tags': None},
u'name': u'stdout',
u'output_type': u'stream',
u'text': u'SMA\nSMA([input_arrays], [timeperiod=30])\n\nSimple Moving Average (Overlap Studies)\n\nInputs:\n price: (any ndarray)\nParameters:\n timeperiod: 30\nOutputs:\n real\n17\n\nWorking on SMA predictions\nparameters used : 808 269\nPerforming stepwise search to minimize aic\n ARIMA(1,3,1)(0,0,0)[0] : AIC=-15000.708, Time=9.21 sec\n ARIMA(0,3,0)(0,0,0)[0] : AIC=-13492.284, Time=2.24 sec\n ARIMA(1,3,0)(0,0,0)[0] : AIC=-15827.971, Time=8.03 sec\n ARIMA(0,3,1)(0,0,0)[0] : AIC=-13635.197, Time=10.27 sec\n ARIMA(2,3,0)(0,0,0)[0] : AIC=-14132.778, Time=3.58 sec\n ARIMA(2,3,1)(0,0,0)[0] : AIC=-15140.312, Time=9.97 sec\n ARIMA(1,3,0)(0,0,0)[0] intercept : AIC=-13970.469, Time=7.12 sec\n\nBest model: ARIMA(1,3,0)(0,0,0)[0] \nTotal fit time: 50.431 seconds\n SARIMAX Results \n==============================================================================\nDep. Variable: y No. Observations: 808\nModel: SARIMAX(1, 3, 0) Log Likelihood 7936.985\nDate: Sun, 12 Dec 2021 AIC -15827.971\nTime: 15:53:06 BIC -15720.081\nSample: 0 HQIC -15786.537\n - 808 \nCovariance Type: opg \n==============================================================================\n coef std err z P>|z| [0.025 0.975]\n------------------------------------------------------------------------------\nx1 -4.786e-05 0.001 -0.066 0.947 -0.001 0.001\nx2 -4.789e-05 0.001 -0.085 0.932 -0.001 0.001\nx3 -4.819e-05 0.000 -0.105 0.917 -0.001 0.001\nx4 1.0000 0.001 1557.248 0.000 0.999 1.001\nx5 -4.579e-05 0.001 -0.071 0.943 -0.001 0.001\nx6 -5.16e-05 0.000 -0.432 0.666 -0.000 0.000\nx7 -4.778e-05 0.000 -0.278 0.781 -0.000 0.000\nx8 -0.0012 0.000 -7.403 0.000 -0.002 -0.001\nx9 -3.454e-06 0.002 -0.002 0.998 -0.003 0.003\nx10 -0.0005 0.001 -0.403 0.687 -0.003 0.002\nx11 0.0029 0.000 10.904 0.000 0.002 0.003\nx12 -0.0003 0.000 -1.815 0.069 -0.001 2.06e-05\nx13 -4.809e-05 0.000 -0.157 0.875 -0.001 0.001\nx14 -0.0001 0.000 -0.482 0.630 -0.001 0.000\nx15 -5.214e-05 0.000 -0.273 0.785 -0.000 0.000\nx16 -4.468e-05 0.000 -0.125 0.901 -0.001 0.001\nx17 -4.224e-05 0.000 -0.202 0.840 -0.000 0.000\nx18 -8.086e-05 0.000 -0.270 0.787 -0.001 0.001\nx19 -5.537e-05 0.000 -0.244 0.807 -0.000 0.000\nx20 8.423e-05 0.000 0.333 0.739 -0.000 0.001\nx21 -4.232e-05 0.000 -0.166 0.868 -0.001 0.000\nar.L1 -0.6666 6.03e-06 -1.11e+05 0.000 -0.667 -0.667\nsigma2 4.093e-10 8.97e-11 4.563 0.000 2.33e-10 5.85e-10\n===================================================================================\nLjung-Box (L1) (Q): 60.24 Jarque-Bera (JB): 1334882.31\nProb(Q): 0.00 Prob(JB): 0.00\nHeteroskedasticity (H): 0.11 Skew: -3.81\nProb(H) (two-sided): 0.00 Kurtosis: 202.35\n===================================================================================\n\nWarnings:\n[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n[2] Covariance matrix is singular or near-singular, with condition number 5.73e+20. Standard errors may be unstable.\nARIMA order: (1, 3, 0) \n\nEpoch 1/500\n\nEpoch 00001: val_loss improved from inf to 0.04321, saving model to LSTM8.h5\n48/48 - 3s - loss: 1.3590 - val_loss: 0.0432 - lr: 0.0010 - 3s/epoch - 69ms/step\nEpoch 2/500\n\nEpoch 00002: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 1.2809 - val_loss: 0.0457 - lr: 0.0010 - 215ms/epoch - 4ms/step\nEpoch 3/500\n\nEpoch 00003: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 1.2036 - val_loss: 0.0497 - lr: 0.0010 - 228ms/epoch - 5ms/step\nEpoch 4/500\n\nEpoch 00004: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 1.1161 - val_loss: 0.0545 - lr: 0.0010 - 221ms/epoch - 5ms/step\nEpoch 5/500\n\nEpoch 00005: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 1.0402 - val_loss: 0.0591 - lr: 0.0010 - 223ms/epoch - 5ms/step\nEpoch 6/500\n\nEpoch 00006: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.\n\nEpoch 00006: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9788 - val_loss: 0.0637 - lr: 0.0010 - 226ms/epoch - 5ms/step\nEpoch 7/500\n\nEpoch 00007: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9461 - val_loss: 0.0642 - lr: 1.0000e-04 - 228ms/epoch - 5ms/step\nEpoch 8/500\n\nEpoch 00008: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9414 - val_loss: 0.0646 - lr: 1.0000e-04 - 215ms/epoch - 4ms/step\nEpoch 9/500\n\nEpoch 00009: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9368 - val_loss: 0.0651 - lr: 1.0000e-04 - 227ms/epoch - 5ms/step\nEpoch 10/500\n\nEpoch 00010: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9323 - val_loss: 0.0656 - lr: 1.0000e-04 - 215ms/epoch - 4ms/step\nEpoch 11/500\n\nEpoch 00011: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.\n\nEpoch 00011: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9278 - val_loss: 0.0662 - lr: 1.0000e-04 - 218ms/epoch - 5ms/step\nEpoch 12/500\n\nEpoch 00012: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9250 - val_loss: 0.0662 - lr: 1.0000e-05 - 225ms/epoch - 5ms/step\nEpoch 13/500\n\nEpoch 00013: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9246 - val_loss: 0.0663 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step\nEpoch 14/500\n\nEpoch 00014: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9242 - val_loss: 0.0663 - lr: 1.0000e-05 - 235ms/epoch - 5ms/step\nEpoch 15/500\n\nEpoch 00015: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9237 - val_loss: 0.0664 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step\nEpoch 16/500\n\nEpoch 00016: ReduceLROnPlateau reducing learning rate to 1e-05.\n\nEpoch 00016: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9233 - val_loss: 0.0664 - lr: 1.0000e-05 - 234ms/epoch - 5ms/step\nEpoch 17/500\n\nEpoch 00017: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9228 - val_loss: 0.0665 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step\nEpoch 18/500\n\nEpoch 00018: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9224 - val_loss: 0.0666 - lr: 1.0000e-05 - 227ms/epoch - 5ms/step\nEpoch 19/500\n\nEpoch 00019: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9219 - val_loss: 0.0666 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step\nEpoch 20/500\n\nEpoch 00020: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9215 - val_loss: 0.0667 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step\nEpoch 21/500\n\nEpoch 00021: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9210 - val_loss: 0.0668 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step\nEpoch 22/500\n\nEpoch 00022: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9206 - val_loss: 0.0668 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step\nEpoch 23/500\n\nEpoch 00023: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9201 - val_loss: 0.0669 - lr: 1.0000e-05 - 236ms/epoch - 5ms/step\nEpoch 24/500\n\nEpoch 00024: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9197 - val_loss: 0.0670 - lr: 1.0000e-05 - 229ms/epoch - 5ms/step\nEpoch 25/500\n\nEpoch 00025: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9192 - val_loss: 0.0670 - lr: 1.0000e-05 - 231ms/epoch - 5ms/step\nEpoch 26/500\n\nEpoch 00026: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9188 - val_loss: 0.0671 - lr: 1.0000e-05 - 215ms/epoch - 4ms/step\nEpoch 27/500\n\nEpoch 00027: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9183 - val_loss: 0.0672 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step\nEpoch 28/500\n\nEpoch 00028: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9179 - val_loss: 0.0673 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step\nEpoch 29/500\n\nEpoch 00029: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9174 - val_loss: 0.0673 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step\nEpoch 30/500\n\nEpoch 00030: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9170 - val_loss: 0.0674 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step\nEpoch 31/500\n\nEpoch 00031: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9165 - val_loss: 0.0675 - lr: 1.0000e-05 - 227ms/epoch - 5ms/step\nEpoch 32/500\n\nEpoch 00032: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9161 - val_loss: 0.0676 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step\nEpoch 33/500\n\nEpoch 00033: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9156 - val_loss: 0.0676 - lr: 1.0000e-05 - 234ms/epoch - 5ms/step\nEpoch 34/500\n\nEpoch 00034: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9152 - val_loss: 0.0677 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step\nEpoch 35/500\n\nEpoch 00035: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9147 - val_loss: 0.0678 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step\nEpoch 36/500\n\nEpoch 00036: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9143 - val_loss: 0.0679 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step\nEpoch 37/500\n\nEpoch 00037: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9138 - val_loss: 0.0679 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step\nEpoch 38/500\n\nEpoch 00038: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9134 - val_loss: 0.0680 - lr: 1.0000e-05 - 216ms/epoch - 5ms/step\nEpoch 39/500\n\nEpoch 00039: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9129 - val_loss: 0.0681 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step\nEpoch 40/500\n\nEpoch 00040: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9125 - val_loss: 0.0682 - lr: 1.0000e-05 - 214ms/epoch - 4ms/step\nEpoch 41/500\n\nEpoch 00041: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9120 - val_loss: 0.0683 - lr: 1.0000e-05 - 224ms/epoch - 5ms/step\nEpoch 42/500\n\nEpoch 00042: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9115 - val_loss: 0.0684 - lr: 1.0000e-05 - 241ms/epoch - 5ms/step\nEpoch 43/500\n\nEpoch 00043: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9111 - val_loss: 0.0684 - lr: 1.0000e-05 - 223ms/epoch - 5ms/step\nEpoch 44/500\n\nEpoch 00044: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9106 - val_loss: 0.0685 - lr: 1.0000e-05 - 221ms/epoch - 5ms/step\nEpoch 45/500\n\nEpoch 00045: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9102 - val_loss: 0.0686 - lr: 1.0000e-05 - 226ms/epoch - 5ms/step\nEpoch 46/500\n\nEpoch 00046: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9098 - val_loss: 0.0687 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step\nEpoch 47/500\n\nEpoch 00047: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9093 - val_loss: 0.0688 - lr: 1.0000e-05 - 222ms/epoch - 5ms/step\nEpoch 48/500\n\nEpoch 00048: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9089 - val_loss: 0.0689 - lr: 1.0000e-05 - 220ms/epoch - 5ms/step\nEpoch 49/500\n\nEpoch 00049: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9084 - val_loss: 0.0690 - lr: 1.0000e-05 - 219ms/epoch - 5ms/step\nEpoch 50/500\n\nEpoch 00050: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9080 - val_loss: 0.0691 - lr: 1.0000e-05 - 218ms/epoch - 5ms/step\nEpoch 51/500\n\nEpoch 00051: val_loss did not improve from 0.04321\n48/48 - 0s - loss: 0.9075 - val_loss: 0.0691 - lr: 1.0000e-05 - 217ms/epoch - 5ms/step\nEpoch 00051: early stopping\n'}
[NbConvertApp] Writing 15784963 bytes to LSTM_Hybrid_using_TA_LIB_Google_Trends.html
Out[93]:
In [ ]: